Cochrane: Library

Cochrane
Library
Cochrane Database of Systematic Reviews

Physical activity and exercise for chronic pain in adults: an overview
of Cochrane Reviews (Review)
Geneen LJ, Moore RA, Clarke C, Martin D, Colvin LA, Smith BH
Geneen LJ, Moore RA, Clarke C, Martin D, Colvin LA, Smith BH.

Physical activity and exercise for chronic pain in adults: an overview of Cochrane Reviews.
Cochrane Database of Systematic Reviews 2017, Issue 4. Art. No.: CD011279.
DOI: 10.1002/14651858.CD011279.pub3.
www.cochranelibrary.com

Physical activity and exercise for chronic pain in adults: an overview of Cochrane Reviews (Review)
Copyright © 2020 The Authors. Cochrane Database of Systematic Reviews published by John Wiley & Sons, Ltd. on
behalf of The Cochrane Collaboration.
Cochrane Trusted evidence.
Informed decisions.

Library Better health. Cochrane Database of Systematic Reviews
TABLE OF CONTENTS
HEADER......................................................................................................................................................................................................... 1
ABSTRACT..................................................................................................................................................................................................... 1
PLAIN LANGUAGE SUMMARY....................................................................................................................................................................... 2
BACKGROUND.............................................................................................................................................................................................. 4
OBJECTIVES.................................................................................................................................................................................................. 5
METHODS..................................................................................................................................................................................................... 5
RESULTS........................................................................................................................................................................................................ 7
Figure 1.................................................................................................................................................................................................. 9
DISCUSSION.................................................................................................................................................................................................. 16
AUTHORS' CONCLUSIONS........................................................................................................................................................................... 18
ACKNOWLEDGEMENTS................................................................................................................................................................................ 19
REFERENCES................................................................................................................................................................................................ 21
ADDITIONAL TABLES.................................................................................................................................................................................... 28
WHAT'S NEW................................................................................................................................................................................................. 65
HISTORY........................................................................................................................................................................................................ 65
CONTRIBUTIONS OF AUTHORS................................................................................................................................................................... 65
DECLARATIONS OF INTEREST..................................................................................................................................................................... 66
SOURCES OF SUPPORT............................................................................................................................................................................... 66
NOTES........................................................................................................................................................................................................... 66
INDEX TERMS............................................................................................................................................................................................... 66
Physical activity and exercise for chronic pain in adults: an overview of Cochrane Reviews (Review) i
Copyright © 2020 The Authors. Cochrane Database of Systematic Reviews published by John Wiley & Sons, Ltd. on behalf of The Cochrane
Collaboration.
Informed decisions.

[Overview of Reviews]
Physical activity and exercise for chronic pain in adults: an overview of

Cochrane Reviews
Louise J Geneen1, R Andrew Moore2, Clare Clarke3, Denis Martin4, Lesley A Colvin5, Blair H Smith6
1Oxford, UK. 2Plymouth, UK. 3Division of Population Health Sciences, University of Dundee, Dundee, UK. 4Institute of Health and
Social Care, Teesside University, Middlesbrough, UK. 5Anaesthesia & Pain Medicine, University of Edinburgh, Western General Hospital,
Edinburgh, UK. 6Division of Population Health Sciences, University of Dundee, Dundee, UK
Contact address: Louise J Geneen, Oxford, UK. louisegeneen@gmail.com.
Editorial group: Cochrane Pain, Palliative and Supportive Care Group

Publication status and date: Edited (no change to conclusions), published in Issue 2, 2020.
Citation: Geneen LJ, Moore RA, Clarke C, Martin D, Colvin LA, Smith BH. Physical activity and exercise for chronic pain in
adults: an overview of Cochrane Reviews. Cochrane Database of Systematic Reviews 2017, Issue 4. Art. No.: CD011279. DOI:
10.1002/14651858.CD011279.pub3.
Copyright © 2020 The Authors. Cochrane Database of Systematic Reviews published by John Wiley & Sons, Ltd. on behalf of The
Cochrane Collaboration. This is an open access article under the terms of the Creative Commons Attribution Licence, which permits use,
distribution and reproduction in any medium, provided the original work is properly cited.
ABSTRACT
Background
Chronic pain is defined as pain lasting beyond normal tissue healing time, generally taken to be 12 weeks. It contributes to disability,
anxiety, depression, sleep disturbances, poor quality of life, and healthcare costs. Chronic pain has a weighted mean prevalence in adults
of 20%.
For many years, the treatment choice for chronic pain included recommendations for rest and inactivity. However, exercise may have
specific benefits in reducing the severity of chronic pain, as well as more general benefits associated with improved overall physical and
mental health, and physical functioning.
Physical activity and exercise programmes are increasingly being promoted and offered in various healthcare systems, and for a variety of
chronic pain conditions. It is therefore important at this stage to establish the efficacy and safety of these programmes, and furthermore
to address the critical factors that determine their success or failure.
Objectives
To provide an overview of Cochrane Reviews of adults with chronic pain to determine (1) the effectiveness of different physical activity and
exercise interventions in reducing pain severity and its impact on function, quality of life, and healthcare use; and (2) the evidence for any
adverse effects or harm associated with physical activity and exercise interventions.
Methods
We searched theCochrane Database of Systematic Reviews (CDSR) on the Cochrane Library (CDSR 2016, Issue 1) for systematic reviews of
randomised controlled trials (RCTs), after which we tracked any included reviews for updates, and tracked protocols in case of full review
publication until an arbitrary cut-off date of 21 March 2016 (CDSR 2016, Issue 3). We assessed the methodological quality of the reviews
using the AMSTAR tool, and also planned to analyse data for each painful condition based on quality of the evidence.
We extracted data for (1) self-reported pain severity, (2) physical function (objectively or subjectively measured), (3) psychological function,
(4) quality of life, (5) adherence to the prescribed intervention, (6) healthcare use/attendance, (7) adverse events, and (8) death.
Due to the limited data available, we were unable to directly compare and analyse interventions, and have instead reported the evidence
qualitatively.
Physical activity and exercise for chronic pain in adults: an overview of Cochrane Reviews (Review) 1
Collaboration.
Informed decisions.

Main results
We included 21 reviews with 381 included studies and 37,143 participants. Of these, 264 studies (19,642 participants) examined exercise
versus no exercise/minimal intervention in adults with chronic pain and were used in the qualitative analysis.
Pain conditions included rheumatoid arthritis, osteoarthritis, fibromyalgia, low back pain, intermittent claudication, dysmenorrhoea,
mechanical neck disorder, spinal cord injury, postpolio syndrome, and patellofemoral pain. None of the reviews assessed 'chronic pain'
or 'chronic widespread pain' as a general term or specific condition. Interventions included aerobic, strength, flexibility, range of motion,
and core or balance training programmes, as well as yoga, Pilates, and tai chi.
Reviews were well performed and reported (based on AMSTAR), and included studies had acceptable risk of bias (with inadequate reporting
of attrition and reporting biases). However the quality of evidence was low due to participant numbers (most included studies had fewer
than 50 participants in total), length of intervention and follow-up (rarely assessed beyond three to six months). We pooled the results
from relevant reviews where appropriate, though results should be interpreted with caution due to the low quality evidence.
Pain severity: several reviews noted favourable results from exercise: only three reviews that reported pain severity found no statistically
significant changes in usual or mean pain from any intervention. However, results were inconsistent across interventions and follow-up,
as exercise did not consistently bring about a change (positive or negative) in self-reported pain scores at any single point.
Physical function: was the most commonly reported outcome measure. Physical function was significantly improved as a result of the
intervention in 14 reviews, though even these statistically significant results had only small-to-moderate effect sizes (only one review
reported large effect sizes).
Psychological function and quality of life: had variable results: results were either favourable to exercise (generally small and moderate
effect size, with two reviews reporting significant, large effect sizes for quality of life), or showed no difference between groups. There were
no negative effects.
Adherence to the prescribed intervention: could not be assessed in any review. However, risk of withdrawal/dropout was slightly higher
in the exercising group (82.8/1000 participants versus 81/1000 participants), though the group difference was non-significant.
Healthcare use/attendance: was not reported in any review.
Adverse events, potential harm, and death: only 25% of included studies (across 18 reviews) actively reported adverse events. Based on
the available evidence, most adverse events were increased soreness or muscle pain, which reportedly subsided after a few weeks of the
intervention. Only one review reported death separately to other adverse events: the intervention was protective against death (based on
the available evidence), though did not reach statistical significance.
Authors' conclusions
The quality of the evidence examining physical activity and exercise for chronic pain is low. This is largely due to small sample sizes and
potentially underpowered studies. A number of studies had adequately long interventions, but planned follow-up was limited to less than
one year in all but six reviews.
There were some favourable effects in reduction in pain severity and improved physical function, though these were mostly of small-to-
moderate effect, and were not consistent across the reviews. There were variable effects for psychological function and quality of life.
The available evidence suggests physical activity and exercise is an intervention with few adverse events that may improve pain severity
and physical function, and consequent quality of life. However, further research is required and should focus on increasing participant
numbers, including participants with a broader spectrum of pain severity, and lengthening both the intervention itself, and the follow-
up period.
PLAIN LANGUAGE SUMMARY
Physical activity and exercise for chronic pain in adults - an overview of Cochrane Reviews
Background
Chronic (long-term) pain is pain that has lasted beyond the body's usual healing time. It is often described as pain that has lasted for at least
three months. Chronic pain causes many problems, beyond the pain itself, including fatigue, anxiety, depression, and a poor quality of life.
In the past, people with chronic pain were told to rest. However, general advice now is to keep active - whether to affect the pain directly
or to combat the other problems associated with it. Therefore, research studies have attempted to examine the effect of physical activity
in people with chronic pain.
This overview aimed to bring together and analyse any reviews published by Cochrane that looked at physical activity and exercise studies
in any chronic pain condition, including arthritis, back and neck pain, and menstrual (period) pain.
Collaboration.
Informed decisions.

Key results and quality of the evidence
In January 2016, we identified 21 Cochrane Reviews which covered 10 different diagnoses (osteoarthritis (a joint disease), rheumatoid
arthritis (joint pain and swelling), fibromyalgia (widespread pain condition), low back pain, intermittent claudication (cramping pain in the
legs), dysmenorrhoea (period pain), mechanical neck disorders (neck pain), spinal cord injury, postpolio syndrome (a condition occurring
in people who have had polio), patellofemoral pain (pain at the front of the knee)). The physical activity or exercise programme used in the
trials ranged in frequency, intensity, and type, including land- and water-based activities, those focusing on building strength, endurance,
flexibility and range of motion, and muscle activation exercises.
The quality of the evidence was low. This was mostly due to the small numbers of people with chronic pain who participated in each
reviewed study. Ideally, a study should have hundreds of people assigned to each group, whereas most of the studies included in the review
process here had fewer than 50 people in total.
There was evidence that physical activity reduced the severity of pain, improved physical function, and had a variable effect on both
psychological function and quality of life. However, these results were not found in all studies. The inconsistency could be due to the quality
of the studies or because of the mix of different types of physical activity tested in the studies. Additionally, participants had predominantly
mild-to-moderate pain, not moderate-to-severe pain.
Conclusions
According to the available evidence (only 25% of included studies reported on possible harm or injury from the intervention), physical
activity did not cause harm. Muscle soreness that sometimes occurs with starting a new exercise subsided as the participants adapted to
the new activities. This is important as it shows physical activity in general is acceptable and unlikely to cause harm in people with chronic
pain, many of whom may have previously feared it would increase their pain further.
Future studies should focus on increasing participant numbers, including a wider range of severity of pain (more people with more severe
pain), and lengthening both the intervention (exercise programme) itself, and the follow-up period. This pain is chronic in nature, and so
a long-term intervention, with longer periods of recovery or follow-up, may be more effective.
Collaboration.
Informed decisions.

BACKGROUND recommendations on the use of exercise, based on evidence drawn

from randomised controlled trials (RCTs), stating: "exercise and
Description of the condition exercise therapies, regardless of their form, are recommended in
the management of patients with chronic pain" (SIGN 2013).
Chronic pain has been defined as pain lasting beyond normal
tissue healing time, generally taken to be 12 weeks (International Description of the interventions
Association for the Study of Chronic Pain; Merskey 2011).
It contributes to disability, anxiety and depression, sleep Physical activity has been defined by the WHO as "any
disturbances, poor quality of life, and healthcare costs (Leadley bodily movement produced by skeletal muscles that requires
2014; Moore 2014a; Park 2012). energy expenditure, including activities undertaken while working,
playing, carrying out household chores, travelling, and engaging
Chronic pain has a weighted mean prevalence in adults of 20% in recreational pursuits" (WHO 2015). WHO also states that
(Breivik 2006; Moore 2014a), which increases as the population "exercise ... is a sub-category of physical activity that is planned,
ages (32% of adults aged 25 to 34 years, 62% of adults over 75 structured, repetitive, and aims to improve or maintain one or more
years; Abdulla 2013; Elliott 1999). This is a greater proportion than components of physical fitness" (WHO 2015).
people with asthma (To 2012) or diabetes (IDF 2012) in the same
population (van Hecke 2013a). The World Health Organization Physical activity for health can take many different forms: it can
(WHO) recognises chronic pain as a public health problem be structured exercise, such as in classes, gym-based, or a DVD
throughout the world, with one systematic review assessing the or programme performed at home; or unstructured and involve
growing evidence that the prevalence of chronic pain in the general adding just a few small activities each day (activities of daily living).
population is high internationally (34% in low-income countries Physical activity and exercise can also vary in intensity, duration,
and 30% in high-income countries; Elzahaf 2012). Chronic painful and type: aerobic (such as walking) or more focused on increasing
conditions comprise four of the 10 highest ranking conditions for flexibility, strength, or balance. Physical activity and exercise can
years lived with disability in 2013 (Vos 2015), and are responsible for also be taught (or led) by another individual such as an exercise
considerable loss of quality of life and employment, and increased professional, or initiated and maintained through the person's own
healthcare costs (Moore 2014b). Despite this, the term 'chronic initiative and motivation.
pain' was only added as a MeSH term in MEDLINE in January 2012
(National Library of Medicine), highlighting the relatively small Both physical activity and exercise can be performed on land or in
proportion of specific research dedicated to this population. the water, and can range from whole-body to localised (body site-
specific) training. Most forms of exercise can also be modified to
Certain factors can contribute to an increased risk of chronic be performed where there is restricted movement (e.g. in a chair, a
pain (female gender, older age, lower socioeconomic status, bed, or another assistive device).
geographical and cultural background, and genetics; Smith 2007;
van Hecke 2013b). Other factors associated with chronic pain How the intervention might work
conditions are modifiable, such as smoking status, alcohol Physical activity and exercise can be adapted for an individual, and
intake, nutrition, obesity, comorbidities, employment status and is something people can do to help themselves. It is likely to be
occupational factors, and physical activity level (Smith 2007; van associated with minimal adverse effects, such as interactions with
Hecke 2013a). medication and potential for abuse in adults with chronic pain,
when compared to pharmaceutical and surgical interventions. It is
A review of current issues in the treatment of chronic pain
therefore an attractive option to help manage an individual's pain
strongly suggests that health professionals traditionally focus
if the systematic reviews show benefit. However, current evidence
on biomedical views of pain, utilising pharmacology first
suggests that simply giving an individual advice to exercise
and foremost, and sometimes not addressing potential non-
is insufficient to bring about significant change (SIGN 2013),
pharmacological approaches such as physical activity and
and a badly prescribed intervention that does not consider the
changing attitudes towards chronic pain (Schofield 2011).
individual's conditions and present state of health and fitness, such
Guidance often suggests that lifestyle advice is important: for
as one that does not incorporate pacing or gradual progression,
example, the National Institute for Health and Care Excellence
may bring about adverse events such as pain 'flare-ups', or lead to
(NICE) osteoarthritis guidelines state that "exercise should be a
cardiac or respiratory events (American College of Sports Medicine
core treatment ... irrespective of age, comorbidity, pain severity
2007). This suggests that supervised or structured interventions
and disability. Exercise should include: local muscle strengthening
may be more fruitful, though this is currently unconfirmed.
[and] general aerobic fitness" (NICE 2014).
Since the 1980s, primary care physician advice for treating
Non-pharmacological treatments have been developed,
pain has changed, moving away from "rest", to minimising or
investigated, and implemented, with Cochrane Reviews and
eliminating bedrest and instead remaining active (back pain,
protocols evaluating the available evidence for psychological,
Waddell 1987). Exercise may have specific benefits in reducing
physical, and other non-medical interventions (e.g. cognitive
the severity of chronic pain, as well as more general benefits
behavioural and behavioural therapy, Eccleston 2014; Williams
associated with improved overall physical and mental health, and
2012; TENS, Nnoaham 2008; low-impact/intensity movement/
physical functioning of people with chronic pain, as depression
exercise therapy, Wieland 2013; dietary, Straube 2015; and patient
(Finan 2013), deconditioning (Bousema 2007), and obesity are
education, Engers 2008; Gross 2009). While evidence for the
commonly observed in these people (headache/migraine, Bigal
effectiveness of these interventions is of variable quantity and
2012; fibromyalgia, Ursini 2011). For example, studies have
quality, the 2013 Scottish Intercollegiate Guideline Network (SIGN)
revealed that a single bout of exercise increases the production of
guidelines on the management of chronic pain made strong
Collaboration.
Informed decisions.

endogenous opioids, leading to transient anti-nociception in both OBJECTIVES

animals and humans, and repeated exercise produces long-lasting
anti-nociception in otherwise untreated animals (Stagg 2011). To provide an overview of Cochrane Reviews of adults with chronic
Aerobic exercise is also strongly linked to weight loss (Messier pain to determine (1) the effectiveness of different physical activity
2013), which in turn has implications for the management of and exercise interventions in reducing pain severity and its impact
chronic pain as the pressure on joints is reduced. Alternatively, on function, quality of life, and healthcare use; and (2) the evidence
resistance exercise, or other forms of strength training, can improve for any adverse effects or harm associated with physical activity
the person's capacity to support bone and cartilage through and exercise interventions.
improved musculature supporting movement around a joint, with
potential to relieve stiffness (Mayer 2008) and bringing about some METHODS
pain relief. Resistance training through repetitive full range-of-
motion exercise around the lumbar spine (in chronic low back Criteria for considering reviews for inclusion
pain) may affect disc metabolism itself, with the possibility that We included only systematic reviews of RCTs of physical activity
the exercise programme could improve metabolic exchange in the and exercise in participants with chronic pain, and published in
lumbar discs and aid in repair (Mooney 2006). Training to improve the Cochrane Database of Systematic Reviews. The included reviews
balance and flexibility also has benefits as it reduces the risk of falls, had to fulfil the following criteria:
and the potential for further pain or injury (Harvard 2013).
Participants
Why it is important to do this overview
Adults (aged 18 years and over) reporting chronic non-cancer pain,
If physical activity and exercise interventions are shown to including persistent (e.g. chronic back pain, fibromyalgia) and
effectively and safely reduce pain intensity or frequency (or both), intermittent (e.g. migraine, dysmenorrhoea) pain, for at least three
they are likely to be a preferable alternative or adjunct therapy months (12 weeks) in any body site.
to pharmacological/surgical treatments for chronic pain. The
interventions could promote personal involvement of individuals Intervention
in the management of their pain, thus increasing self-efficacy
Reviews of RCTs assessing physical activity or exercise as the
and the ability to self-manage. In turn this could lead to an
intervention (any reviews where that assessed physical activity
increase in overall quality of life and a consequent reduction in
or exercise as a stand-alone intervention). This included physical
healthcare use. In addition, exercise is of great importance for
activity interventions that could be initially taught by an exercise
cardiovascular (Vigorito 2014) and bone health (Sakuma 2012).
professional, or involve periodical/ongoing supervision.
Reduced physical function and consequent lack of mobility in
people with chronic pain is associated with increased all-cause and Exclusions
cardiovascular mortality (Nüesch 2011), with other studies linking
severe chronic pain to general increased all-cause mortality (Moore Interventions not deemed physical activity or exercise using the
2014a; Torrance 2010). WHO definition, such as manipulation, mobilisation, or passive
movement. Any multi-modal interventions were excluded if
Physical activity and exercise programmes are increasingly being physical activity/exercise could not be assessed for effect (the effect
promoted and offered in various healthcare systems (American of exercise must have been measured distinctly).
College of Sports Medicine (ACSM) 'Exercise is Medicine' global
pledge at the Inaugural World Congress 2010) and for a variety Comparison
of chronic pain conditions, including arthritis (Fransen 2014; Silva Usual care, waiting list control, placebo/sham treatment, other
2010), fibromyalgia (Busch 2013), and dysmenorrhoea (Brown treatment, or a combination of treatments (as long as the effect of
2010). At this stage it is important to establish the efficacy and exercise could be measured distinctly).
safety of these programmes, and furthermore to address the critical
factors that determine their success or failure. Primary outcome
It is therefore important to identify whether (and how) exercise • self-reported pain (severity).
interventions can be effectively and safely applied in people with
This could be presented and analysed as change on a continuous
chronic pain.
scale, the proportion of participants who 'responded', or, ideally,
With a number of systematic reviews published by Cochrane in a dichotomised format as the proportion of participants in each
evaluating the effectiveness of exercise in various painful group who achieved a predetermined threshold of improvement
conditions, it is timely and important to bring together all relevant (e.g. outcome in individual participants of at least 50% pain
published information to evaluate the current evidence, and intensity reduction, or no worse than mild pain, at the end of the
identify the availability and quality of evidence-based exercise trial, with at least 30% pain intensity reduction as a secondary
interventions. This overview will determine the extent to which outcome, or recovery; Moore 2013).
the published systematic reviews have accurately assessed the
Secondary outcomes
evidence for exercise in chronic pain conditions/syndromes, which
will help to direct future guidelines and identify current research • Physical function (objectively or subjectively measured).
gaps. • Psychological function.
• Quality of life.
• Adherence to the prescribed intervention.
Collaboration.
Informed decisions.

• Healthcare use/attendance. statistical software due to the lack of relevant and comparable data
• Adverse events (not death). (RevMan 2014).
• Death. We collected the following information (where available) from the
Reviews may not always report specifically on activity or reviews:
exercise for chronic pain in adults. We anticipated two possible
• number of included studies and participants;
circumstances which might have arisen.
• intervention (exercise or activity type) and dose (frequency/
• A review included some interventions of interest or reported intensity);
only some outcomes of interest. In this case we extracted the • comparator;
interventions and outcomes of interest, but we did not include • condition treated;
interventions or outcomes outside the scope of this overview. • time of assessment;
• Reviews occasionally included papers that included children • duration of follow-up;
and adults together, but the results for adults were not reported
• relevant outcomes.
or analysed separately in the included papers or the review. In
this case we made a judgement as to whether the review could Where possible we extracted risk ratio (RR), number needed to
be included based on the proportion of adults. Our intention treat for an additional beneficial outcome (NNTB), mean difference
was to include only those reviews where more than 80% of (MD), and standardised mean difference (SMD), and other relevant
participants were adults. statistical data for the primary and secondary outcomes. This
included:
Search methods for identification of reviews
We searched the Cochrane Database of Systematic Reviews (CDSR), • obtaining 50% pain relief (participant-reported);
2016, Issue 1, on the Cochrane Library for relevant reviews using • obtaining any other measure of 'improvement' (participant-
the search strategy: (pain or migraine or headache) and (exercise or reported);
activity or physical). We did not seek non-Cochrane reviews. • adverse events;
• death;
Data collection and analysis
• withdrawals.
Two overview authors (LG, CC) independently carried out searches
and selected reviews for inclusion. Disagreements were resolved Assessment of methodological quality of included reviews
through discussion, and a third overview author (RAM) acted as Quality of included reviews
arbitrator where necessary.
Two overview authors (LG, CC) independently assessed each
Two overview authors (independently carried out assessment of included review to see if it satisfied the criteria specified
methodological quality (LG, CC), and extracted data (LG, RAM). Any in the 'assessment of multiple systematic reviews' (AMSTAR)
disagreements were resolved through discussion, or involving a measurement tool (Shea 2007), for rigorous methodological
third overview author if necessary (DM). quality. Arbitration by a third overview author (DM) was necessary
for some fields.
One overview author (LG) tracked results of the search for the most
up to date version of each review and protocol that fulfilled the High quality reviews were required to fulfil each of the established
inclusion criteria. AMSTAR criteria (further criteria to fulfil each field is listed in Table
1).
Selection of reviews
For each review we also planned to assess the likelihood of
Included reviews assessed RCTs of the effects of exercise for
publication bias by calculating the number of participants in
pain management in adults (as defined by individual reviews),
studies with zero effect (relative benefit of one) that would be
compared with any of the listed comparators, and included:
needed to give an NNTB too high to be clinically relevant (Moore
• a clearly defined clinical question; 2008). In this case we would have considered an NNTB of 10 or
greater for the outcome of participant-reported pain relief of 30% or
• details of inclusion and exclusion criteria;
greater to be the cut-off for clinical relevance. This method is used
• details of databases searched and relevant search strategies; as statistical tests for the presence of publication bias have been
• participant-reported pain severity (primary outcome measure); shown to be unhelpful (Thornton 2000). However, assessment of
• summary results for at least one other desired outcome. publication bias was not possible due to the lack of specificity of the
populations included within the reviews, and so we were unable to
Data extraction and management extract comparable data.
Two overview authors (LG, RAM) independently extracted data from
Quality of evidence in included reviews
the included review using a standardised data extraction form
and checked for agreement prior to entry into Microsoft Excel for We planned to use two main indicators for the quality of evidence:
Windows. We did not extract data from reports included in the all included reviews must have used only primary studies that were
reviews again, neither did we undertake any re-analysis of data both randomised and double-blind, so minimising the risk of bias
from reviews. Data were not entered for analysis into Cochrane's from these items; and all included reviews must have included only
people with at least moderate pain intensity at baseline (visual
Collaboration.
Informed decisions.

analogue scale greater than 30/100, categorical rating scale greater • random sequence generation (selection bias);
than 1/3, and numerical rating scale greater than 3/10, Collins • allocation concealment (selection bias);
1997), providing a sensitive assay of intervention efficacy. • blinding of participants and personnel (performance bias);
Subsequently, we planned to analyse data for each painful • blinding of outcome assessment (detection bias);
condition in three tiers, according to outcome and freedom from • incomplete outcome data (attrition bias);
known sources of bias. • selective reporting (reporting bias);
• sample size;
• The first tier used data meeting current best standards,
where studies reported the outcome of at least 50% pain • any other biases.
intensity reduction from baseline (where 50% was the cut-off
Data synthesis
for a dichotomous (yes/no) outcome: was a 50% reduction
in pain observed?), or its equivalent, without using last Additional quantitative analyses were not required, since we only
observation carried forward (LOCF) or other imputation method considered results from properly conducted (Cochrane) reviews.
for dropouts, reported an intention-to-treat (ITT) analysis, lasted The aim was to concentrate on specific outcomes such as the
eight or more weeks, had a parallel-group design, and had at proportion of participants with at least 50% pain relief, all-cause
least 200 participants (preferably at least 400) in the comparison or adverse event discontinuations, or serious adverse events, and
(Moore 2010). These top-tier results were usually reported first. to explore how these can be compared across different treatments
• The second tier used any available data, but where one or more for the same condition. We planned to compare only like with like
of these conditions were not met, for example reporting at (where possible); for example in study duration, which can be an
least 30% pain intensity reduction, using LOCF or a completer additional source of bias if insufficient in length (Moore 2010).
analysis, lasting four to eight weeks, and where the numbers of
participants were at least 200. However due to the limited data available, we were unable to
directly compare and analyse interventions, and have instead
• A third tier of evidence related to small amounts of data (fewer
reported the evidence qualitatively only. We had also planned
than 200 participants), or short studies of less than four weeks,
to employ subgroup analyses assessing age, condition, and
or where there was obvious major heterogeneity between
intervention type/intensity, though this was not feasible using the
studies, or where there were other shortcomings in allocation
available data from included reviews. For this reason we have also
concealment, considerable attrition, and incomplete outcome
been unable to include a 'Summary of findings' table as planned
data. For this third tier of evidence, no data synthesis was
and stated in the protocol.
reasonable, and may have been misleading, but an indication of
beneficial effects might be possible. Importantly, we have tried to highlight issues of low trial quality,
inadequate size, and whether trials were truly valid for the
This overview examined the quality of all included reviews
particular condition in making between-therapy comparisons.
according to current best standards for reporting in pain. These
included the attempt and ability of the reviews to identify studies/ We approached each review with four main questions/focus, and
interventions with the maximum evidence of effectiveness, and extracted data accordingly.
minimum risk of bias, including the reporting of the following.
• Did they report exercise versus non-exercise studies?
• Outcomes in trials of the proportion of participants obtaining at
• Did the review or studies included in the review (or both) have
least 50% pain intensity reduction, or no worse than mild pain,
low risk of bias?
at the end of the trial (with at least 30% pain intensity reduction
as a secondary outcome). We did not consider the use of mean • Did they have our main outcome?
changes in pain scores as high quality because responses to pain • What were the actual intervention/s included in the review?
interventions are not Gaussian, and few people have the mean
response. RESULTS
• Duration of included studies of eight weeks or longer. We included 21 reviews with 381 included studies, totalling 37,143
• Imputation method of baseline observation carried forward participants. Of these, 264 studies (19,642 participants) examined
(BOCF), LOCF, or worst observation carried forward (WOCF) if exercise versus no exercise/minimal intervention in adults with
adverse event withdrawals were similar in active and control chronic pain (the focus of this overview) and so were used in the
groups. qualitative analysis.
• At least 200 participants per treatment group in included
studies, with at least two trials, as a minimum criterion for Description of included reviews
trustworthiness of any analysis. Pooled analysis of small studies The search strategy was performed in the Cochrane Library only,
may be considered good quality if at least 400 participants were and revealed 475 potentially relevant titles, of which 75 were
involved, but we regarded these as being potentially subject to assessed as full papers.
bias.
The search was undertaken on 31 January 2016 (CDSR 2016, Issue
We extracted the 'Risk of bias' as assessed by the original review 1), after which any included reviews were tracked for updates, and
authors from included reviews. Counts of low risk of bias were protocols were followed in case of full review publication until 21
extracted from relevant studies in the included reviews and March 2016 (CDSR 2016, Issue 3).
tabulated under the following headings to evaluate the proportion
of studies achieving a low risk of bias for each:
Collaboration.
Informed decisions.

All extracted data and methodological quality assessment were included once published as a full review, one which was unclear,
taken from the most recent published version of the full review. and four that were excluded based on information within the
protocol). Hence, we excluded 54 titles (10 protocols and 44 full
Ultimately, of the 75 titles requiring further assessment, 10 were reviews; Figure 1), reasons for which are listed in Table 2.
reviews at protocol stage only (five of which have potential to be

Collaboration.
Informed decisions.

Figure 1. Study flow diagram.

Collaboration.
Informed decisions.

Figure 1. (Continued)

Detailed information about the included reviews is available in Aquatic exercise
Table 3. Trial and participant number, age, and gender distribution
Any exercise performed in water. This can include swimming,
is reported in Table 4.
though many studies will be referring to exercises performed
Specificity of chronic pain condition of included reviews vertically in the water (not horizontally), either using the water to
support the body through the exercise, or as resistance against the
Following abstract and full paper assessment, 21 reviews fulfilled body.
the inclusion criteria: four in rheumatoid arthritis (Cramp 2013;
Han 2004; Hurkmans 2009; Silva 2010), four in osteoarthritis Range of motion and flexibility exercise
(Bartels 2007; Fransen 2014; Fransen 2015; Regnaux 2015), three
Can be performed in water or on land. The intention is to increase
in fibromyalgia (Bidonde 2014; Busch 2007; Busch 2013), three
the range of motion around a joint through progressive stretching
in low back pain (Hayden 2005; Saragiotto 2016; Yamato 2015),
and mobilising of the muscles around and crossing the joint. For
two in intermittent claudication (Lane 2014; Lauret 2014), one in
the purposes of this overview, we only included active movement
dysmenorrhoea (Brown 2010), one in mechanical neck disorder
where the movement was brought about by the participant, and
(Gross 2015a), one in spinal cord injury (Boldt 2014), one in
not passively moved by an external force such as a therapist.
postpolio syndrome (Koopman 2015), and one in patellofemoral
pain (van der Heijden 2015). None of the included reviews assessed Aerobic exercise
'chronic pain' or 'chronic widespread pain' as a general term or
specific condition. Can be performed in water or on land. Exercise usually performed
continuously to raise the heart rate and breathing rate for a
The 21 included reviews were published by five different Cochrane prolonged period. Examples include walking, jogging, running,
Review groups: 11 from the Cochrane Musculoskeletal Group cycling, and swimming. Often presented as a percentage of the
(Bartels 2007; Bidonde 2014; Busch 2007; Busch 2013; Cramp 2013; participant's heart rate max (HRmax) - the highest heart rate
Fransen 2014; Fransen 2015; Han 2004; Hurkmans 2009; Regnaux reached when performing at their absolute maximum. Similarly
2015; Silva 2010); four from the Cochrane Neck and Back Group it may be presented as a percentage of VO2max or VO2peak (a
previously the Cochrane Back Group) (Gross 2015a; Hayden 2005; proportion of the maximum amount of oxygen the muscle can take
Saragiotto 2016; Yamato 2015); two from the Cochrane Peripheral up per minute), or as an absolute value (mL/kg/minute).
Vascular Diseases Group (Lane 2014; Lauret 2014); one from the
Cochrane Menstrual Disorders and Subfertility Group (Brown 2010); Strength/resistance exercise
one from the Cochrane Injuries Group (Boldt 2014); one from the
Can be performed in water or on land. Exercise performed against
Cochrane Neuromuscular Group (Koopman 2015); and one from
a progressive resistance with the intention of improving muscle
the Cochrane Bone, Joint and Muscle Trauma Group (van der
strength, muscle endurance, muscle power, or a combination of
Heijden 2015).
these. Resistance can come from fixed or free weights, elastic
Protocols that may be included in updates of this overview focus bands, body weight (against gravity), and water resistance. It may
on osteoarthritis (Østerås 2013 from the Cochrane Musculoskeletal also involve static or isometric strength (holding a position or
Group), migraine (Brønfort 2015 from the Cochrane Pain, Palliative weight without moving against it). Often presented as a percentage
and Supportive Care Group), chronic low back pain (Hayden of the participant's one repetition maximum (1-RM) - the maximum
2012 from the Cochrane Back Group), ankylosing spondylitis weight they can lift/move if they only have to do it once.
(Regnaux 2014 from the Cochrane Musculoskeletal Group), and
Motor control exercise
temporomandibular disorders (Craane 2006 from the Cochrane
Oral Health Group). Can be performed in water or on land. Exercise to bring about
activation of the deep trunk muscles, targeting the restoration of
Exercise and physical activity interventions implemented in control and co-ordination of these 'core muscles' (Saragiotto 2016).
the included reviews
Balance (proprioceptive) training
Interventions assessed included: any specified style of land-based
exercise or physical activity such as one designed to improve Can be performed in water or on land (water may be used initially
strength, range of movement, aerobic capacity, or a combination for support). Exercise emphasises the maintenance of balance
of these (Boldt 2014; Busch 2007; Busch 2013; Cramp 2013; Fransen during visual and perturbation challenges with eyes open or closed,
2014; Fransen 2015; Gross 2015a; Hurkmans 2009; Koopman 2015; range of motion, and maintaining stability over reduced areas of
Regnaux 2015; van der Heijden 2015); a single style of land-based support and unstable surface (Silva 2010), that is improving balance
exercise only (tai chi only: Han 2004, walking only: Lauret 2014, in increasingly unstable situations.
walking or jogging only: Brown 2010; Lane 2014, balance training
only: Silva 2010, motor control exercise only: Saragiotto 2016, Tai chi
Pilates method only: Yamato 2015); any pool-based or aquatic An ancient Chinese discipline developed from martial arts,
therapy (Bartels 2007; Bidonde 2014; Cramp 2013), or "any exercise involving a continuous series of very controlled (and usually slow)
therapy" (Hayden 2005). movements designed to improve physical and mental wellbeing.
Collaboration.
Informed decisions.

Yoga week × number of weeks), for a more accurate and detailed

analysis.
Arising out of Hindu philosophy. Exercise includes breath control,
simple meditation, and the adoption of specific bodily postures. Intervention specificity for chronic pain in the included
It is widely practised for health, relaxation, and control (physically reviews
and mentally). Incorporates stretching and flexibility training
with isometric strength training (holding certain poses, with no The focus of this overview was exercise versus no-exercise
movement against a resistance). interventions with the intention of answering the original question:
is exercise beneficial, detrimental, or ineffective for people with
Pilates chronic pain when compared to inactivity? Two of the 21 reviews
did not include/locate any studies that examined simply exercise
Developed by Joseph Pilates in the 20th Century, it is a system
versus no exercise (Lauret 2014; Silva 2010). However, many of the
of exercises (often using special apparatus) designed to improve
included reviews compared varying exercise modality, duration,
physical strength, flexibility, and posture, and enhance mental
intensity, and frequency. The "no-exercise" intervention referred to
awareness.
the control group where there was a minimal intervention (such as
Duration and dose (frequency/intensity) of the exercise and sham exercise or education) or wait-list control/no treatment (see
physical activity interventions Table 3 for more information on control group activity).
A detailed breakdown of each review can be seen in Table 5. Time points reported
Duration of intervention Four of the 19 reviews that reported data, reported results at a
single time point only ('post-intervention': Bidonde 2014; Busch
Interventions assessed by the included reviews varied in length 2007; Cramp 2013; Han 2004). Reviews also analysed outcome
from a single session (Fransen 2015) to 30 months (Fransen 2015). measures immediately post-intervention and at one or more
Only five reviews enforced a minimum intervention period to follow-up points. Each review defined short-, intermediate-, and
reduce risk of bias, and were able to attribute any effects to the long-term follow-up according to their own assessment, so when
intervention (Brown 2010; Busch 2013; Gross 2015a; Hurkmans the time period was not mentioned explicitly, we grouped the
2009; Silva 2010). reviews according to the review authors' own classification only,
and where a time period (weeks, month, years) was explicitly
Frequency listed but not defined by the authors, we grouped them as short-
There was large variation in the exercise or physical activity term (follow-up as under six months), intermediate-term (six to
intervention being implemented, ranging from just once a week 12 months), and long-term (longer than 12 months): short-term:
(Bidonde 2014; Busch 2007; Fransen 2014; Fransen 2015; Han Busch 2013; Fransen 2014; Fransen 2015; Gross 2015a; Hayden
2004; Saragiotto 2016), to twice a day (Boldt 2014), and some 2005; Lane 2014; Regnaux 2015; Saragiotto 2016; intermediate-
performing a short series of exercises (two-minute duration) every term: Bartels 2007; Fransen 2015; Gross 2015a; Hayden 2005; Lane
15 minutes during the day (Gross 2015a). However, when reported, 2014; Regnaux 2015; Saragiotto 2016; long-term: Gross 2015a;
most included studies in the reviews implemented the programme Hayden 2005; Regnaux 2015; Saragiotto 2016. Five reviews did not
twice a week (or stated at least twice a week). report "post-intervention" but at short-term, mid/intermediate-
term, and long-term postrandomisation (short, mid, and long term:
Intensity Boldt 2014; short and intermediate term: Koopman 2015; Yamato
2015; short and long-term: Hurkmans 2009; van der Heijden 2015).
Few studies quantified the intensity of each session. Baseline
One review assessed participants in an ongoing fashion "over three
intensity was often accepted as low/moderate, with the aim to
menstrual cycles" (Brown 2010).
progress over the intervention period to 70% to 85% of HRmax
or heart rate reserve (HRR) for aerobic interventions (Brown 2010; Long-term follow-up
Cramp 2013; Hurkmans 2009), 70% to 80% of an individual's 1-
RM, or 50% to 70% maximum voluntary contraction (Koopman Of the seven reviews claiming to report "long term" follow-up, one
2015) in strength/resistance training programmes (Busch 2013; classed long-term as longer than six weeks (intermediate term as
Hurkmans 2009). In other reviews, intensity was described more one to six weeks' follow-up) (Boldt 2014). The remaining six reviews
loosely as "variable" or "low intensity (very light) to maximum defined long-term follow up as over 12 months (one year) post-
effort (vigorous)" (Bidonde 2014; Fransen 2014; Lane 2014; Regnaux intervention (Gross 2015a; Hayden 2005; Hurkmans 2009; Regnaux
2015), "low intensity" (Fransen 2014; Gross 2015a; Han 2004; Silva 2015; Saragiotto 2016; van der Heijden 2015).
2010), or "moderate or moderate-to-high" (Cramp 2013; Fransen
2015). Methodological quality of included reviews
AMSTAR quality assessment of included reviews
Duration (per session)
No review achieved a perfect score of 11/11, though five achieved
Individual sessions varied in length from two minutes (Gross
10/11 (Boldt 2014; Busch 2013; Hayden 2005; Koopman 2015;
2015a), to 90 minutes (Busch 2013; Cramp 2013; Han 2004) or 120
Regnaux 2015) and eight scored 9/11 (Cramp 2013; Gross 2015a;
minutes (Boldt 2014), but mostly situated around 45 to 60 minutes.
Hurkmans 2009; Lane 2014; Lauret 2014; Saragiotto 2016; van der
However, it is important to note that the shorter sessions were
Heijden 2015; Yamato 2015). The lowest score was 6/11 (Silva 2010)
often performed more regularly than longer sessions. With more
though five categories were not applicable (n/a) due to there being
information it would have been possible to calculate total volume
no included studies. Quality assessment results for each individual
of exercise or physical activity (session duration × frequency per
review are presented in Table 6.
Collaboration.
Informed decisions.

All reviews except one (Bidonde 2014) fulfilled the basic criteria exclude these studies unnecessarily from their analysis (Lane 2014;
(questions one to three of Table 1); to follow an 'a priori' design Lauret 2014). Without these two reviews, only a small percentage
as Cochrane implements a system of protocol publication before (7.8% or 18/229) of the included studies would have scored low risk
undertaking the full reviews, where it also specifies dual study of performance bias (blinding of participants and personnel), but by
selection and data extraction from a comprehensive literature including them (all 35 studies from those two reviews assessed as
search. One review did not fulfil the 'a priori' design as this was low risk of bias) the overall proportion of studies assessed as having
an update and separation from a broader review series, and so the low risk of bias was closer to 20% (53/264).
criteria had not been explicitly listed prior to publication for this
specific title (Bidonde 2014). Attrition (incomplete outcome data, withdrawals/dropouts)
About 55% (144/264) of the studies included in these reviews
Criteria which scored badly using the AMSTAR tool were
showed low risk of bias.
characteristics of included studies (question six of Table 1),
reporting of publication bias (question 10 of Table 1), and conflict Reporting bias (selective reporting)
of interest declarations (question 11 of Table 1).
Reporting bias was classed as low risk in only 46% of included
• Included study characteristics were limited, often reporting the studies. However, it is important to note this was not due to the
"inclusion criteria" used to recruit participants in the study remainder having high risk of bias, but instead 'unclear', as trial
instead of the characteristics of actual included participants, protocols were not always published or accessible to the review
and excluding information such as participants' age, gender authors to accurately assess/interpret.
split, ethnicity, and disease status.
• Assessment of publication bias was omitted entirely in five Study/sample/group size
reviews (Bartels 2007; Fransen 2014; Fransen 2015; Han 2004; Sample size was not always included within the risk of bias
Hurkmans 2009), and when it was assessed, it was reported assessment. It was therefore extracted directly from each review's
using only a simple statement (with no test values, analyses table of included study characteristics by a single overview author
used, or diagrams to demonstrate the result; Busch 2007; (LG), and assessed as being low risk of bias when there was a
Koopman 2015). Two reviews mentioned in the methods as minimum of 50 participants per arm, or 100 in total. Numbers were
planned analyses, though was not mentioned again (Brown then separated for the proportion of studies with greater than 100
2010; van der Heijden 2015), and a third review mentioned it participants per arm (or 200 in total), and 200 participants per arm
in the methods, but appeared to use it interchangeably with (or 400 in total), as this could then be considered higher tiered
reporting bias causing great confusion (Bidonde 2014). evidence.
• Conflicts of interest were sufficiently reported in only three
out of 21 of the included reviews (Hayden 2005; Koopman Only 26 out of 264 included studies (10%) across the 21 reviews
2015; Silva 2010). In the remaining reviews, a cursory statement reported over 100 participants in total (or 50 per arm), a further 6%
was commonly made regarding the review authors' conflicts of (15/264) included over 200 participants per arm. The remaining 223
interests, however, fulfilling the AMSTAR criteria also requires a studies (84%) had fewer than 50 participants per arm (or sample
statement to be made regarding any conflict of interest for any size was not reported), often not reaching 50 in total.
of the included studies.
Other bias
Risk of bias in included reviews The format for reporting bias has changed, and therefore some
The original review authors assessed risk of bias (see Table 7). The earlier reviews (that are yet to be updated) did not assess bias using
table shows the number of studies assessed as low risk of bias only, the same format. Others reported additional criteria as 'other bias'
and excluded those that were assessed as unclear or high risk of including the similarity of baseline characteristics, and similarity of
bias. timing points.
Selection bias (randomisation and allocation concealment) Interpretation of results/conclusions by original review
authors
Selection bias had the largest proportion of included studies with
low risk of bias (63% and 42% of studies adequately undertaking For conclusions made by the original review authors, see Table
and reporting the methods used). 8. We assessed whether these conclusions/interpretations of the
results accurately reflected the information provided within the
Performance and detection bias (blinding participants, review, and if any further information should have been included.
personnel, outcome assessors) This final assessment of the review is an important stage in
determining any author bias within the review process, as many
With any exercise or physical activity intervention it is very difficult
readers, funders, and policy makers will focus on the author
to blind both participants and personnel to the allocation, though
conclusions without a full appraisal of the actual presented data.
some studies included in reviews attempted to by offering sham
exercise. Eleven of the 21 reviews reported appropriate conclusions based
on the data available in the context of the quality of evidence
Due to the difficulty of blinding participants to their group
(Bidonde 2014; Boldt 2014; Busch 2007; Busch 2013; Fransen 2015;
allocation, review authors assessed the risk of bias in different
Gross 2015a; Koopman 2015; Regnaux 2015; Saragiotto 2016; Silva
ways, which may cause confusion: whereas the majority declared
2010; Yamato 2015); five reviews had appropriate conclusions,
this lack of possible blinding to be high risk of bias or unclear,
did not mention quality of the evidence in the conclusion, but
two reviews labelled such cases as low risk of bias in order not to
did discuss it in detail earlier in the review (Bartels 2007; Cramp
Collaboration.
Informed decisions.

2013; Han 2004; Hayden 2005; Lauret 2014); two reviews had Two of the 21 reviews did not include/identify any studies that
appropriate conclusions but had only limited discussion of quality examined intervention versus control (Lauret 2014; Silva 2010).
or did not adequately consider the quality of the evidence in the Of the remaining reviews that did report studies examining
interpretation of the results (Hurkmans 2009; Lane 2014); and three intervention versus control (no physical activity or exercise, or
reviews needed further comment as the strength of the conclusions minimal intervention), two did not report pain as an absolute or
were not appropriate based on the available data (Brown 2010; relative score of severity, intensity, or change as a result of the
Fransen 2014), or we were unable to agree with their interpretation intervention (Brown 2010; Han 2004), and one review assessed
due to difficulty in extracting the data (van der Heijden 2015). pain-free time and distance during exercise (they did not assess
pain using a mean/usual pain scale; Lane 2014). We could not
Effect of interventions extract relevant data for one review as they compared two different
exercise interventions and a control but did not report the data
We have interpreted results using data reported in the reviews,
compared to the control (Regnaux 2015).
and did not return to the original studies. Where data have been
reported as MDs or as an absolute or relative change score we The remaining 15 reviews reported a mean or usual pain score
have used the appropriate scales (where possible) to determine for exercise (intervention) and no-exercise (control) groups (Bartels
whether this was clinically significant. When data have only been 2007; Bidonde 2014; Boldt 2014; Busch 2007; Busch 2013; Cramp
presented as SMD, with or without 95% confidence intervals (CI), 2013; Fransen 2014; Fransen 2015; Gross 2015a; Hayden 2005;
with or without level of significance (P value), we have cautiously Hurkmans 2009; Koopman 2015; Saragiotto 2016; van der Heijden
used the interpretation by Cohen 1988 who defined effect size using 2015; Yamato 2015).
the SMD as small (SMD 0.2 to 0.5), moderate (SMD 0.5 to 0.8), or large
(SMD greater than 0.8). Reported baseline pain score
For the purposes of clarity, we have used the term 'intervention' to Of the 15 reviews that were able to assess pain (Table 9), only three
refer to the exercise or physical activity intervention, and 'control' reviews reported actual baseline pain scores (Bidonde 2014; Boldt
to refer to the included comparison group which did not involve any 2014; Hayden 2005). Three reviews reported change data (Bartels
exercise or physical activity element. 2007; Busch 2007; Busch 2013), but we were able to use control
group baseline and earliest control group scores as assumed or
Primary outcome approximate baseline measures for the intervention groups in nine
reviews (Bartels 2007; Busch 2007; Fransen 2014; Fransen 2015;
Self-reported pain (severity)
Gross 2015a; Koopman 2015; Saragiotto 2016; van der Heijden 2015;
Part of the inclusion criteria for this overview was for pain severity Yamato 2015). Overall, only three reviews that assessed pain did
to be listed as an outcome measure. not provide baseline or control group scores for comparison (Busch
2013; Cramp 2013; Hurkmans 2009).

Intervention group at baseline Control group at baseline Control group at earliest follow-up
Median pain score 70.9/100 WOMAC 9.1/20 (2 studies, n = Mean pain score ˜ 29/100
380)
(based on 7 studies, n = 382; Bidonde (9 studies, n = 549; Fransen 2014)
2014) VAS ˜ 55/100 (3 studies, n =
117)
HAQ 1.05/3 (1 study, n = 249)

(Bartels 2007)
11.05 to 22.6 on a 0 to 150 WUSPI VAS 35/100 to 61/100 44/100

score
(4 studies, n = 204; Busch 2007) (44 studies, n = 3537; Fransen 2015)
(1 study, n = 35; Boldt 2014)
Mean pain score 46/100 (95% CI 41 to - 40/100 to 60/100

50)
(2 studies, n = 147; Gross 2015a)
(8 studies, n = 370; Hayden 2005)
- - 44/100 SD 24
(1 study, n = 55; Koopman 2015)
- - range 25/100 to 56/100
(4 studies, n = 291; Saragiotto 2016)
Collaboration.
Informed decisions.

- - 2.1/10 to 6.0/10
(2 studies, n = 41; van der Heijden 2015)
- - range 18/100 to 52/100
(6 studies, n = 148; Yamato 2015)
Range: 46 to 70.9 on a 0 to 100 scale Range: 35 to 55 on a 0 to 100 Range: 18 to 60 on a 0 to 100 scale

scale
16 studies, n = 787 68 studies, n = 4768
10 studies, n = 950
CI: confidence interval; HAQ: Health Assessment Questionnaire; n: number of participants; SD: standard deviation; VAS: visual ana-
logue score; WOMAC: Western Ontario and McMaster Universities Osteoarthritis Index; WUSPI: Wheelchair User's Shoulder Pain Index.
HAQ: mean of different category scores, 0 or 1 (mild to moderate disability), up to 2 or 3 (severe to very severe disability); WOMAC
pain score: 5 items summed to 0 (no pain) to 20 (worst pain ever); WUSPI: 15 items of 0 to 10 VAS scores, summed to form total of 0
(no pain) to 150 (worst pain ever).

This suggests the majority of participants reviewed had mild- changes in usual or mean pain from any intervention (Cramp 2013;
to-moderate pain (only one review reported a mean of severe Hurkmans 2009; Koopman 2015 (assumed due to lack of presented
pain (aquatic exercise for fibromyalgia, Bidonde 2014) at the data)). The remaining reviews reported a statistically significant
commencement of each intervention (less than 30/100 mild pain, effect of the intervention at one or more time points, in at least one
30/100 to 60/100 moderate pain, more than 60/100 severe pain; subgroup.
Collins 1997), though labelling the majority as having only mild-to-
moderate pain should be interpreted with caution due to the lack of Three reviews found at least 30% pain reduction from baseline
specific data available - the baseline data of the intervention group (post-intervention - strength training: Busch 2007; Busch 2013, at
would have been preferable to the proxies we have had to use. short-term follow-up: van der Heijden 2015). Additionally, seven
reviews reported clinically significant results (minimally important
Quality judgement/ tiered quality (first, second, third tier evidence) difference: reduction in pain from baseline of at least 10 points on
Our assessment criteria stated that we would accept the a 0 to 100 scale or an absolute improvement of at least 10% to
information as graded evidence when reported as the number of 20%, Dworkin 2008) as a result of the exercise intervention (1.3/10
participants achieving a 50% (first tier evidence) or 30% (second from aerobic training, Busch 2007; 12/100 (95% CI 10 to 15), Fransen
tier evidence) reduction in pain, but none of the included reviews 2015,; 14.9/100 (95% CI 7.39 to 22.40), Gross 2015a; 10.2/100 (95% CI
reported results in this way, and so instead we used the reported 1.31 to 19.09), Hayden 2005; 2.5/10 (95% CI 1.52 to 3.48), Boldt 2014;
absolute and relative change values. 10.01/100 (95% CI 4.35 to 15.67), Saragiotto 2016; 14.05/100 (95%
CI 9.19 to 18.91), Yamato 2015). Three reviews found statistically
None of the included reviews fulfilled the requirements for first tier significant improvements as a result of the intervention, but they
evidence (at least 50% pain reduction from baseline, study duration did not reach clinical significance (post-intervention, P = 0.02,
longer than eight weeks, and more than 200 participants per arm). Bartels 2007; "small to moderate" benefit post-intervention and at
six-month follow-up, P < 0.001, Fransen 2014; "moderate effect" of
Second tier evidence (at least 30% pain reduction from baseline, 7% (95% CI 3 to 11) benefit post-intervention, Bidonde 2014).
study duration between four and eight weeks, and more than 200
participants in total or 100 participants per arm) was also lacking in Overall, results were inconsistent across interventions and follow-
these reviews; three reviews found at least 30% reduction in pain up (see Table 9), as exercise did not consistently bring about a
from baseline (Busch 2007; Busch 2013; van der Heijden 2015), one change (positive or negative) in self-reported pain scores at any
of which also used long enough exercise programmes (eight to 21 single point.
weeks' intervention, Busch 2013) but totalled only 81 participants
across two studies. The other two reviews did not fulfil the study Secondary outcomes
duration criteria (interventions from 2.5 weeks, Busch 2007; and Physical function (objectively or subjectively measured)
three weeks, van der Heijden 2015) or study size criteria.
Measures of physical function were the primary outcome measure
Consequently results from relevant reviews have been pooled (all in eight out of 21 reviews (Busch 2013; Han 2004; Hayden 2005;
tier three quality) where appropriate, though results should be Hurkmans 2009; Koopman 2015; Lane 2014; Lauret 2014; Silva
interpreted with caution due to the low quality evidence. 2010), and a reported (non-primary) outcome measure in nine more
reviews (Bartels 2007; Bidonde 2014; Busch 2007; Fransen 2014;
Treatment effect Fransen 2015; Gross 2015a; Regnaux 2015; Saragiotto 2016; van der
Data that could be extracted for pain can be seen in Table 9 for Heijden 2015, plus some which assessed disability; Cramp 2013;
all reviews. Only three reviews found no statistically significant Saragiotto 2016; Yamato 2015). Only Boldt 2014 and Brown 2010 did
Collaboration.
Informed decisions.

not list physical function (or disability, or activity limitation) as a 2005). These have been reported separately to quality of life (Table
potential outcome measure. 12).
Treatment effect Treatment effect
Data that could be extracted for physical function are shown in Data that could be extracted for quality of life can be seen in
Table 10. Two reviews which reported physical function had no data Table 12. Four reviews found no significant difference between
to extract (Lauret 2014; Silva 2010), and for one review we were intervention and control groups in health-related quality of
unable to extract the relevant data (Regnaux 2015). Two reviews life post-intervention (9 studies, n = 556) (HRQoL: Boldt 2014;
found no significant difference in physical function between the Fransen 2014; Gross 2015a, global assessment: Bidonde 2014;
intervention and control groups (Han 2004; Hurkmans 2009, both Gross 2015a)), three reviews did not or were unable to report any
rheumatoid arthritis, 8 studies, n = 240). The remaining 14 reviews data (HRQoL: Lauret 2014, global assessment: Hayden 2005, other
showed that the intervention produced a statistically significant assessment: Silva 2010), and seven reviews found a significant
benefit over the control at a minimum of one reported time point improvement as a result of the intervention (34 studies, n = 2700)
(Bartels 2007; Bidonde 2014; Busch 2007; Busch 2013; Cramp 2013; (HRQoL: Bartels 2007, Fransen 2015, global assessment: Busch
Fransen 2014; Fransen 2015; Gross 2015a; Hayden 2005; Koopman 2007; Saragiotto 2016; Yamato 2015, other assessment: Bidonde
2015; Lane 2014; Saragiotto 2016; van der Heijden 2015; Yamato 2014; Busch 2013).
2015; 129 studies, n greater than 9559 (exact number unknown due
to some participant numbers not being reported)). Two reviews assessing strength/resistance training interventions
found significantly large effect sizes (SMD greater than 0.8, as
Many of these statistically significant results were of small or defined by Cohen 1988) in favour of the intervention (global
moderate effect size (as reported by the review authors, or using wellbeing measure, SMD 1.43 (95% CI 0.76 to 2.10), Busch 2007;
the definition by Cohen 1988 if unreported; small effect size: Bartels Fibromyalgia Impact Questionnaire, SMD 1.27 (95% CI 0.72 to 1.83),
2007; Bidonde 2014; Fransen 2014; Fransen 2015; Gross 2015a; Busch 2013). Other statistically significant changes reported in the
Koopman 2015; Saragiotto 2016; Yamato 2015, moderate effect size: included reviews were of small-to-moderate effect size (SMD 0.2 to
Busch 2007; Fransen 2015; Yamato 2015). 0.8, Cohen 1988).
Only one review reported statistical significance and large effect Adherence to the prescribed intervention
size (both short-term and long-term follow-up: SMD 1.10 (95% CI
Only one review reported adherence to the intervention as an
0.58 to 1.63) and 1.62 (95% CI 0.31 to 2.94), van der Heijden 2015).
outcome measure (Regnaux 2015), but the authors were unable
However, the original review authors highlighted the low to very
to perform an analysis on attendance as most studies did not
low quality of the evidence as many studies had high or unclear risk
clearly report attendance or compliance (Regnaux 2015). However,
of bias across multiple domains (van der Heijden 2015).
five reviews assessed withdrawals or dropouts (Bidonde 2014;
Psychological function Fransen 2014; Han 2004; Regnaux 2015; Saragiotto 2016), one
reported all-cause attrition (Busch 2013), and another reported the
Only five out of 21 reviews assessed psychological function as discontinuation rate (Silva 2010).
mental health (Bartels 2007; Bidonde 2014; Busch 2013), anxiety
(Cramp 2013), and depression (Boldt 2014; Busch 2013; Cramp Data that could be extracted for adherence, withdrawals, and
2013). attrition can be seen in Table 13. Pooling all available data for
withdrawals/dropout/attrition gave an RR of 1.02 (95% CI 0.94 to
Treatment effect 1.12) in favour of the control group (6 reviews, 30 studies, n = 2256,
Data that could be extracted for psychological function can be control withdrawal 81/1000, intervention withdrawal 82.8/1000).
seen in Table 11. There were significant effects in favour of
the intervention for mental health (Bartels 2007) and depression One clinically controlled trial (CCT) in one review reported
(Busch 2013) scores, and "variable effect" for depression (Cramp statistically significant improvement in enjoyment of exercise/rest
2013). However, there was also no effect or no differences (P = 0.0002) and self-reported benefit from exercise/rest (P = 0.006)
between control and intervention groups reported for mental at both post-intervention (end of therapy, 10 weeks) and follow-up
health (Bidonde 2014; Busch 2013), anxiety (Cramp 2013), and (four months later) (n = 95, Han 2004).
depression (Boldt 2014). Healthcare use/attendance
Quality of life None of the reviews reported healthcare use/attendance.
A version of quality of life assessment was reported in nine reviews. Adverse events (not death)
Six were termed quality of life or health-related quality of life
(HRQoL) (Bartels 2007; Boldt 2014; Fransen 2014; Fransen 2015; Eighteen out of 21 reviews reported adverse effects (three reviews
Gross 2015a; Lauret 2014). did not report adverse events as an outcome measure due to
lack of studies or other undisclosed reasons; Brown 2010; Lauret
Other reviews assessed global perceived effect (Gross 2015a), 2014; Silva 2010). Two reviews only assessed a specific adverse
global wellbeing (Busch 2007), global assessment (Hayden 2005), event ("amputation" Lane 2014; "motor unit survival" Koopman
global impression of recovery (Saragiotto 2016; Yamato 2015), 2015), one review observed "safety - pain and radiological
health assessment questionnaire (Silva 2010), multi-dimensional damage" (Hurkmans 2009), and another referred to any "side-
function (Bidonde 2014; Busch 2013), and work status (Hayden effects" (Han 2004).
Collaboration.
Informed decisions.

Data that could be extracted for adverse events (not death) can large effect sizes for quality of life), or showed no difference
be seen in Table 14. The total number of reported adverse events between groups. There were no negative effects.
(not death) was 137 events across 39 studies out of 61 studies
that had adverse events as an outcome measure (over one-third Adherence to the prescribed intervention: could not be assessed
of all trials that reported them found no adverse events related to in any included review. However, risk of withdrawal/dropout was
the intervention): six reviews reported no adverse events from the slightly higher in the exercising group (82.8/1000 participants
included trials (Bartels 2007; Busch 2013; Cramp 2013; Hurkmans versus 81/1000 participants), though the group difference was not
2009; Koopman 2015; Yamato 2015) though the authors questioned significant.
whether this was due to lack of reporting by the trial authors, or
Healthcare use/attendance: not reported in any included review.
whether there were no adverse events.
Adverse events, potential harm, and death: importantly, exercise
Adverse events were largely reported as a total number per trial,
caused no actual harm, with most adverse events being increased
though one review separately reported results for the intervention
soreness or muscle pain, which reportedly subsided after several
group versus the control group (Saragiotto 2016), and two others
weeks of the intervention. One review reported a non-significant
reported adverse events for the intervention group only (Boldt
reduction in risk of death as a result of the intervention.
2014; Regnaux 2015). Only one review calculated an RR for the
adverse events, showing a reduced risk for amputation in the
Overall completeness and applicability of evidence
intervention group (two amputations in the usual care/control
group: RR 0.20, 95% CI 0.01 to 4.15, based on one study in one Of the 21 included reviews, seven could be considered out of date as
review, Lane 2014). they were most recently assessed as up-to-date prior to 2010 such
that any recent controlled trials assessing pain severity have not
Death been included in this overview (Cochrane recommends updating
Only one out of 21 reviews reported death separately to other reviews every two years) (Bartels 2007; Brown 2010; Busch 2007;
adverse events (Lane 2014). Based on five studies within the review, Han 2004; Hayden 2005; Hurkmans 2009; Silva 2010). We included
death had an RR of 0.71 (95% CI 0.28 to 1.78) in favour of exercise as these reviews in the overview, but they may not be as relevant now
being protective, though was not statistically significant (P = 0.47). due to the elapsed time since they were updated. One protocol that
had potential to be included was published in 2006 with no full
DISCUSSION review available yet (Craane 2006).
Specificity of the condition: despite the heterogeneous nature Available data suggest that participants in the included reviews and
of chronic pain, in this overview we have combined several studies would generally be characterised as having mild-moderate
painful conditions covering a number of conditions and diagnoses. pain (moderate greater than 30/100 or 3/10) with only one review
Regardless of aetiology, the impact of chronic pain is broadly reporting moderate-severe pain (severe greater than 60/100 or
similar across many conditions. 6/10). Therefore whether the evidence of change or no change seen
here as a result of each intervention is applicable to people further
Summary of main results along on the pain spectrum (with higher pain scores/worse pain) is
debatable. However, it can be argued that those people are more
Pain severity: there were favourable results in a number of reviews likely to be assigned medical or surgical interventions than physical
as a result of exercise: only three reviews found no statistically activity and exercise alone (where available), and as a group they
significant changes in usual or mean pain from any intervention. may be less able to engage in exercise, and may therefore be
However, results were inconsistent across interventions and follow- more difficult to recruit into exercise-only studies. Having said this,
up, as the intervention did not consistently bring about a change the labelling of participants as having mild-moderate pain was
(positive or negative) in self-reported pain scores at any single a cautious one within this overview due to the lack of specific
point. The exercise or physical activity interventions did not have a data available at baseline assessment; only three reviews included
negative effect on the outcome (did not worsen the pain). A factor baseline pain scores in the intervention group, and two further
in the lack of statistical and clinically significant result may be the reviews provided control group baseline scores.
baseline pain severity of participants. The majority of the included
population had an assumed mild-to-moderate pain severity score There are still gaps in the available literature, and therefore also
(assumed only due to lack of exact group data at baseline). This within this overview. None of the included reviews examined
is often the desired outcome (post-intervention) of many drug generalised or widespread chronic pain as a global condition, each
therapies for pain, and it may therefore be difficult to show a instead examined specific conditions that included chronic pain
clinically significant improvement in these people. as a symptom or result of the ongoing condition (rheumatoid
arthritis, osteoarthritis, fibromyalgia, low back pain, intermittent
Physical function: physical function/disability was the most claudication, dysmenorrhoea, mechanical neck disorder, spinal
commonly reported outcome measure, and was the primary cord injury, postpolio syndrome, and patellofemoral pain). The pain
measure in eight out of the 21 reviews. Physical function was in these cases can occur secondary to other symptoms such as
significantly (statistically) improved as a result of the intervention fatigue, muscle stiffness, difficulty sleeping, and depression, all of
in 14 reviews, though even these statistically significant results had which could separately (and more effectively) be influenced by the
only small-to-moderate effect sizes in all but one review. intervention. Additionally, only 25% of included studies actively
reported adverse events. This may affect the completeness of the
Psychological function and quality of life: there were variable
evidence as conclusions have been drawn based on the available
results for psychological function and quality of life: results were
data. The included reviews did not discuss the possible impact
either favourable to exercise (two reviews reporting significantly
Collaboration.
Informed decisions.

of this non-reporting by the original trials, and this may lead to in future, to adequately assess the influence of small trials on
underestimating possible adverse events from an intervention, or the estimated treatment effect (Nüesch 2010). Inclusion in the
overestimating its safety. standard assessment process may in turn influence the design and
undertaking of future research trials to increase the sample size,
The exercise interventions examined in the included reviews were and produce more consistent clinically and statistically accurate
broad; including aerobic, strength, flexibility, range of motion, and results.
core or balance training programmes, as well as yoga, Pilates,
and tai chi. Many of these interventions can be accessed in the Of the 21 included reviews, 12 used a pain measure as their
community by the general public and people with chronic pain, primary outcome (Bartels 2007; Boldt 2014; Brown 2010; Busch
either individually or in classes (yoga, Pilates, tai chi). Other 2007; Fransen 2014; Fransen 2015; Gross 2015a; Hayden 2005;
exercise intervention programmes, such as the motor control Regnaux 2015; Saragiotto 2016; van der Heijden 2015; Yamato
exercise and proprioceptive (balance) training, requires at least 2015), and the remaining nine reviews included the measure as a
initial supervision by a therapist to teach the correct techniques secondary outcome only. Other outcomes were shared, including
and provide feedback for progression. physical and psychological function, and quality of life. Likewise,
each review team will have included studies that did not use their
Quality of the evidence chosen outcome measures as the primary measure, and that were
therefore powered according to a different primary outcome. On
In assessing the quality of the evidence, we employed the AMSTAR
collating the evidence, some studies may appear underpowered for
tool to examine the reviews, extracted data on risk of bias
the outcome(s) of interest to us (Turner 2013), yet were adequately
to examine the available primary evidence, and evaluated the
powered for the studies' primary measure. To increase the power
authors' conclusions to ensure that they were appropriate based
of the results of this overview, and the intermediary reviews we
on the available data.
have included, intervention studies that focus on painful conditions
The AMSTAR tool is useful in assessing the reporting of a systematic should include pain intensity as the primary outcome, or at least
review, though it does not inform us of the actual undertaking as a prominent secondary outcome; alternatively review authors
or conduct of the review process. All 21 included reviews scored should seek to include only those studies that were adequately
well across the AMSTAR assessment, though this is likely due powered for pain intensity as a primary outcome measure.
to the stringent reporting guidelines implemented by Cochrane
Intervention length ranged from a single session to regular
prior to publication. However, it may be necessary or advisable
sessions over a period of 30 months, though the majority were
for the Cochrane guidelines to be further expanded and detailed
between eight and 12 weeks. Durations of this length are common
with regards to reporting study characteristics, publication bias,
among exercise and physical activity intervention studies to allow
and conflicts of interest, as these areas often did not meet the
for physiological adaptation and familiarisation. In contrast, the
requirements laid out in the AMSTAR criteria (Table 1).
follow-up period was often inadequate, as many reviews reported
Data extracted from the reviews regarding their assessment of only a single follow-up point (immediately post-intervention), or
bias (risk of bias) showed moderate level scores at best across repeated measures over the short-term (less than six months):
all included studies within the included reviews. Other than only six of the 21 reviews planned to assess participants over
issues surrounding blinding (which are problematic in exercise the long term (over 12 months: Gross 2015a; Hayden 2005;
intervention studies due to the nature of the intervention), the trials Hurkmans 2009; Regnaux 2015; Saragiotto 2016; van der Heijden
did not consistently and adequately report potential attrition and 2015). With chronic conditions, it would be advisable to include
reporting biases, with less than half of studies within these reviews longer follow-up periods (beyond 12 months post-randomisation)
at low risk of bias. as long-term solutions may be more relevant to their control or
pain management. It is also possible that initial adaptation and
However, the most prominent issue with regards to bias in these potential benefits as a result of an exercise intervention may take
exercise and physical activity intervention studies is the sample size longer to manifest in comparison to a 'healthy' person due to the
used. This subcategory is not used as standard in the assessment possible limitations in exercise intensity and progression (a training
of bias in Cochrane Reviews, despite the increasing volume of threshold) beyond which any additional physical training may
research available suggesting that small studies of fewer than 100 be detrimental to the underlying pathophysiological mechanisms
participants per arm (Moore 2010; Nüesch 2010) are at increased (Daenen 2015) or simply be additional physical stress with no
risk of succumbing to the random effects in estimating both additional physical benefit (Benton 2011).
direction and magnitude of treatment effects (Moore 1998; Turner
2013) due to greater heterogeneity within and between small We grouped outcome measurement points in this overview into
studies (IntHout 2015). short term (less than six months), intermediate term (six to 12
months), and long term (longer than 12 months). The broad
Studies within the included reviews here were very small (often time window for 'short term' outcomes (less than six months)
fewer than 50 participants in total). For greater quality and a more is a potential source of heterogeneity as the early period is
reliable effect, at least 100 participants per arm should be analysed the one where time of measurement is most likely to result in
for a study to potentially be classed as tier two evidence (200 per variable outcomes. These initial problems could be overcome by
arm for tier one); small studies are known to overestimate the use of standard reporting periods in exercise intervention studies
treatment effect by up to 32% in comparison with larger studies (suggested four-weekly within the 'short term' period to assess
(Deschartes 2013). both neural adaptation and other physiological changes). This
would allow review authors to use the data gathered closest to
Assessing studies for risk of bias based on study size (total number the time point they are assessing, for more accurate analyses.
or per arm) should be included in any review or meta-analysis
Collaboration.
Informed decisions.

Additionally, by extending the follow-up period beyond one year Implications for practice
(long-term follow-up), heterogeneity may be reduced further.
For clinicians and people with chronic pain
Reviews generally did not enforce a minimum exercise requirement The evidence in this overview suggests that the broad spectrum of
for inclusion in their review. Additionally, not all exercise sessions physical activity and exercise interventions assessed here (aerobic,
were supervised or baseline fitness/physical ability was assessed strength, flexibility, range of motion, and core or balance training
subjectively, and consequently it was not reported whether the programmes, as well as yoga, Pilates, and tai chi) are potentially
intervention was fulfilled as described, or whether the dose was beneficial, though the evidence for benefit is low quality and
enough to elicit a physiological response. Studies often rely on inconsistent. The most commonly reported adverse events were
the self-report of participants as to the actual physical activity increased soreness or muscle pain, which subsided after several
and exercise being undertaken, which can lead to a greater risk weeks of the intervention.
of bias, and reduced study quality as it is questionable as to
whether the effect can be truly attributed to the intervention. This Physical activity and exercise may improve pain severity as well as
was examined in a previous review, where it was concluded that physical function and quality of life.
non-subjective physical assessment should be performed where
possible (Perruchoud 2014), though these still have challenges For policy makers
regarding implementation.
The evidence showed variable results, though in some reviews
In summary, the quality of the evidence was low (third tier): within there was a clinical and statistical benefit in pain relief and physical
this overview we found no tier one or tier two evidence. This is function (based on low quality evidence). The evidence suggests
largely due to the small sample sizes and potentially underpowered that physical activity or exercise is an acceptable intervention in
studies. A number of studies within the reviews had adequately people with chronic pain, with minimal negative adverse effects.
long interventions, but planned follow-up was limited to less than However based on this low quality evidence, we cannot provide
one year (12 months) in all but six reviews. direction to the content of an exercise programme should clinicians
decide to implement one.
Interpretation of the available data, and conclusions drawn by
the review authors, were appropriate, although the conclusions Implications for research
were sometimes stronger than warranted by the available data. There is a clear need for further research into exercise and physical
Occasionally results were not discussed with regards to the quality activity for chronic pain in adults.
of the evidence or risk of bias: it is important to discuss the
findings in the context of the quality of the evidence, with complete General implications
transparency, as this may affect future research, and implications
for patients, funders, and policy makers. • Future research should report baseline values for outcome
measures in both intervention and control groups, together with
Potential biases in the overview process detailed relevant information about the participants. Knowing
the baseline value is relevant to interpreting any change
While we have attempted to include all relevant reviews in the observed as a result of the intervention, and understanding the
overview process, we do concede that by only searching the broader value of the intervention.
Cochrane Library, and including only current Cochrane Reviews • Where possible, pain results should be reported as the number
we may have missed some key literature. However previous of people achieving 50%, 30%, and 10% pain relief, and the
publications have referred to the higher quality grading (high number who did not meet that point (dichotomous outcome).
AMSTAR score) in Cochrane Reviews due to the basic criteria These are clinically important cut-offs in pain intervention
necessary for publication at any stage (protocol or full review) research, and reporting in this way allows readers to observe the
suggesting they may be the most reliable source of evidence clinical effect more effectively.
(O'Connell 2013).
• Reporting should include median and range as well as mean
Agreements and disagreements with other studies or and standard deviation (SD) of results. This will allow readers
reviews to review the effects of any outliers that may have skewed the
data, which often goes unnoticed in the reporting of mean and
This is a summary overview of current Cochrane Reviews, we are SD alone.
not aware of any overviews or reviews summarising non-Cochrane • The importance of clear intervention reporting is
reviews. underestimated: often studies report both intervention
and control programmes simply, where other researchers
AUTHORS' CONCLUSIONS and clinicians alike are unable to replicate the trial or
intervention. Recommendations for reporting are based on
There is limited evidence of improvement in pain severity as a
the Consolidated Standards of Reporting Trials (CONSORT)
result of exercise. There is some evidence of improved physical
statement (www.consort-statement.org/), but this alone
function and a variable effect on both psychological function and
does not detail the extent of necessary intervention and
quality of life. However, results are inconsistent and the evidence is
control programmes reporting. The template for intervention
low quality (tier three). Promisingly however, none of the physical
description and replication (TIDieR) approach (Hoffman 2014) is
and activity interventions assessed appeared to cause harm to the
intended as an extension to CONSORT item 5 ("The interventions
participants.
for each group with sufficient details to allow replication,
including how and when they were actually administered")
Collaboration.
Informed decisions.

and is a checklist for detailing the programmes using: why these important restrictions can be implemented as subgroup
(rationale), what (materials and procedures), who, how, where, analyses, though if this is the case it is important to have
when, and how much. adequate study numbers (ideally 200 participants per arm or
subgroup).
Design • Due to the chronicity and long-term nature of the condition,
• One previous review highlighted the increased bias often physiological and psychological changes may take longer to
present in questionnaires and other self-report measures of manifest. It is widely accepted that there is a delay in muscular
physical activity in people with chronic pain, and as a result hypertrophy as a result of exercise, and initial gains within the
made the recommendation to use objective measures instead, first few weeks of any training programme will be as a result of
such as accelerometers, or the use of direct and indirect neural factors (Enoka 1997); this is also in line with the grading
calorimetry, where possible (Perruchoud 2014), though these of evidence (tier two evidence or higher requires a minimum of a
still have challenges regarding implementation. This would four-week intervention). This suggests that longer interventions
allow direct and exact comparison and analyses of actual energy may be necessary (eight weeks for tier one evidence), though
expenditure and treatment effect. assessing participants at regular intervals, including at four
weeks, would be beneficial to examine the effect of the neural
Population/participants/sample adaptation alone.
• There needs to be a focus on participants with generalised and/ Measurement (end-points)
or widespread chronic pain, instead of (or as well as) condition-
specific populations. • Randomised controlled trials with long-term follow-up are
• Studies should include people with higher pain severity (greater needed. Chronic pain is defined by its chronic nature, and
than 50/100 on a 100-point visual analogue scale) at baseline. therefore long-term follow-up of results is equally important as
People with mild-moderate pain should still be included, but it the initial short-term effect (if not more so): outcomes should
would be advisable to separate the results for analysis, ensuring be assessed beyond one year after randomisation. In turn this
the study is adequately powered to allow this subgroup analysis will inform the direct effect of the intervention, as well as the
in advance. This way we could determine if exercise has benefit proportion of the population who maintains the programme of
overall, or affects one group more than another, and tailor exercise employed in the intervention, or something else under
exercise programmes according to the individual needs. the guise of physical activity as a result of participation.
• It has been previously suggested that for 20% to 25% of • The broad time window for 'short term' outcomes (less than
participants undertaking an exercise programme there is little six months) is a potential source of heterogeneity as the early
to no favourable response (Timmons 2014), while a small period is the one where time of measurement is most likely
percentage (5% to 10%) have adverse events (Bouchard 2012). It to result in variable outcomes. These initial problems could
is therefore vitally important that much larger sample sizes are be overcome by use of standard reporting periods in exercise
used: ideally more than 200 participants per arm, though even intervention studies (suggested four-weekly assessment within
this number in total would increase the quality of the evidence in the 'short term' period to assess both neural adaptation and
the first instance. In this way we may be able to learn to identify other physiological changes). This would allow review authors
individuals who will benefit, and those who will require further to use the data recorded closest to the time point they are
intervention. assessing, for more accurate and comparable analyses.
• Outcome measures used by researchers should be standardised
Interventions across trials and studies. Recommendations for selecting the
most appropriate and important outcome measures to those
• Different forms of exercise should be researched in detail. For
who live with chronic pain have previously been published
the purposes of this overview, we combined all physical activity
(Initiatives on Methods, Measurement, and Pain Assessment
and exercise interventions under one banner to determine if
in Clinical Trials (IMMPACT) Consensus Recommendations,
there was any effect. However a number of reviews separately
Dworkin 2005; Turk 2003).
analysed resistance (strength) training, aerobic (endurance),
and combination programmes. It is important to continue to Other
examine different modalities, but currently there is not enough
high quality evidence to exclude or prioritise one specific mode • It would be of interest in future research to determine the
(resistance, endurance, stability) or medium (land/water based), reasons for non-participation in regular physical activity or non-
or the proportion of a combination programme to be assigned to compliance to a prescribed exercise intervention in people with
each, as all may have individual benefits for people with chronic chronic pain, and how to overcome these barriers.
pain. • Future Cochrane Reviews could include: exercise for chronic
• Intensity of exercise, duration of individual sessions, and pain or chronic widespread pain (and not specific conditions
frequency should be investigated. It is this dose alongside such as osteoarthritis, fibromyalgia, etc.), and exercise for
duration (of the entire intervention) and adherence that may neuropathic pain. These areas have not been covered by
determine the actual efficacy. Cochrane with an exercise or physical activity intervention.
• More reviews and trials should attempt to minimise intervention
heterogeneity by implementing minimum and maximum
ACKNOWLEDGEMENTS
requirements. Only this way will the research community be Thanks to Dr Amanda Williams, Professor Chris Eccleston, and Anna
able to determine more accurately the direction and magnitude Erskine at the Cochrane Pain, Palliative and Supportive Care Review
of effect of a specific programme or intervention. Many of Group (PaPaS), for their support and rigour in the development of
Collaboration.
Informed decisions.

the concept of this overview, title proposal, and protocol. Thanks Disclaimer: the views and opinions expressed therein are those
also to peer reviewers Dr Neil O'Connell and Dr Clare Stevinson, and of the authors and do not necessarily reflect those of the NIHR,
consumer referees Jennifer Patrick and Dr Shailen Mehta. National Health Service (NHS), or the Department of Health.
Cochrane Review Group funding acknowledgement: the National External funding support from the National Institute of Academic
Institute for Health Research (NIHR) is the largest single funder of Anaesthesia (BJA RCoA NIAA) and the MRC Lifelong Health and
the Cochrane PaPaS Group. Wellbeing Grant to complete the overview as part of two larger
projects.
Collaboration.
Informed decisions.

REFERENCES

References to included reviews Cochrane Database of Systematic Reviews 2015, Issue 1. [DOI:
10.1002/14651858.CD004376.pub3]
Bartels 2007
Bartels EM, Lund H, Hagen KB, Dagfinrud H, Christensen R, Gross 2015a
Danneskiold-Samsøe B. Aquatic exercise for the treatment of Gross A, Kay TM, Paquin JP, Blanchette S, Lalonde P,
knee and hip osteoarthritis. Cochrane Database of Systematic Christie T, et al. Exercises for mechanical neck disorders.
Reviews 2007, Issue 4. [DOI: 10.1002/14651858.CD005523.pub2] Cochrane Database of Systematic Reviews 2015, Issue 1. [DOI:
10.1002/14651858.CD004250.pub5]
Bidonde 2014
Bidonde J, Busch AJ, Webber SC, Schachter CL, Danyliw A, Han 2004
Overend TJ, et al. Aquatic exercise training for fibromyalgia. Han A, Judd M, Welch V, Wu T, Tugwell P, Wells GA. Tai chi for
Cochrane Database of Systematic Reviews 2014, Issue 10. [DOI: treating rheumatoid arthritis. Cochrane Database of Systematic
10.1002/14651858.CD011336] Reviews 2004, Issue 3. [DOI: 10.1002/14651858.CD004849]
Boldt 2014 Hayden 2005
Boldt I, Eriks-Hoogland I, Brinkhof MWG, de Bie R, Joggi D, Hayden J, van Tulder MW, Malmivaara A, Koes BW. Exercise
von Elm E. Non-pharmacological interventions for therapy for treatment of non-specific low back pain.
chronic pain in people with spinal cord injury. Cochrane Cochrane Database of Systematic Reviews 2005, Issue 3. [DOI:
Database of Systematic Reviews 2014, Issue 11. [DOI: 10.1002/14651858.CD000335.pub2]
10.1002/14651858.CD009177.pub2]
Hurkmans 2009
Brown 2010
Hurkmans E, van der Giesen FJ, Vliet Vlieland TPM, Schoones J,
Brown J, Brown S. Exercise for dysmenorrhoea. Cochrane Van den Ende ECHM. Dynamic exercise programs (aerobic
Database of Systematic Reviews 2010, Issue 2. [DOI: capacity and/or muscle strength training) in patients with
10.1002/14651858.CD004142.pub2] rheumatoid arthritis. Cochrane Database of Systematic Reviews
2009, Issue 4. [DOI: 10.1002/14651858.CD006853.pub2]
Busch 2007
Busch AJ, Barber KA, Overend TJ, Peloso PMJ, Schachter CL. Kay 2012
Exercise for treating fibromyalgia syndrome. Cochrane Kay TM, Gross A, Goldsmith CH, Rutherford S, Voth S,
Database of Systematic Reviews 2007, Issue 4. [DOI: Hoving JL, et al. Exercises for mechanical neck disorders.
10.1002/14651858.CD003786.pub2] Cochrane Database of Systematic Reviews 2012, Issue 8. [DOI:
10.1002/14651858.CD004250.pub4]
Busch 2013
Busch AJ, Webber SC, Richards RS, Bidonde J, Schachter CL, Koopman 2015
Schafer LA, et al. Resistance exercise training for fibromyalgia. Koopman FS, Beelen A, Gilhus NE, de Visser M,
Cochrane Database of Systematic Reviews 2013, Issue 12. [DOI: Nollet F. Treatment for postpolio syndrome. Cochrane
10.1002/14651858.CD010884] Database of Systematic Reviews 2015, Issue 5. [DOI:
10.1002/14651858.CD007818.pub3]
Cramp 2013
Cramp F, Hewlett S, Almeida C, Kirwan JR, Choy EHS, Chalder T, Lane 2014
et al. Non-pharmacological interventions for fatigue in Lane R, Ellis B, Watson L, Leng GC. Exercise for intermittent
rheumatoid arthritis. Cochrane Database of Systematic Reviews claudication. Cochrane Database of Systematic Reviews 2014,
2013, Issue 8. [DOI: 10.1002/14651858.CD008322.pub2] Issue 7. [DOI: 10.1002/14651858.CD000990.pub3]
Fransen 2008 Lauret 2014
Fransen M, McConnell S. Exercise for osteoarthritis of the knee. Lauret GJ, Fakhry F, Fokkenrood HJP, Hunink MGM, Teijink JAW,
Cochrane Database of Systematic Reviews 2008, Issue 4. [DOI: Spronk S. Modes of exercise training for intermittent
10.1002/14651858.CD004376.pub2] claudication. Cochrane Database of Systematic Reviews 2014,
Issue 7. [DOI: 10.1002/14651858.CD009638.pub2]
Fransen 2014
Fransen M, McConnell S, Hernandez-Molina G, Regnaux 2015
Reichenbach S. Exercise for osteoarthritis of the hip. Regnaux JP, Lefevre-Colau MM, Trinquart L, Nguyen C,
Cochrane Database of Systematic Reviews 2014, Issue 4. [DOI: Boutron I, Brosseau L, et al. High-intensity versus low-intensity
10.1002/14651858.CD007912.pub2] physical activity or exercise in people with hip or knee
osteoarthritis. Cochrane Database of Systematic Reviews 2015,
Fransen 2015
Issue 10. [DOI: 10.1002/14651858.CD010203.pub2]
Fransen M, McConnell S, Harmer AR, Van der Esch M,
Simic M, Bennell KL. Exercise for osteoarthritis of the knee.
Collaboration.
Informed decisions.

Saragiotto 2016 Craane 2006

Saragiotto BT, Maher CG, Yamato TP, Costa LOP, Menezes Craane B, De Laat A, Dijkstra PU, Stappaerts K, Stegenga B.
Costa LC, Ostelo RWJG, et al. Motor control exercise for chronic Physical therapy for the management of patients with
non-specific low-back pain. Cochrane Database of Systematic temporomandibular disorders and related pain. Cochrane
Reviews 2016, Issue 1. [DOI: 10.1002/14651858.CD012004] Database of Systematic Reviews 2006, Issue 1. [DOI:
10.1002/14651858.CD005621]
Silva 2010
Silva KNG, Mizusaki Imoto A, Almeida GJM, Atallah ÁN, Dagfinrud 2008
Peccin MS, Fernandes Moça Trevisani V. Balance training Dagfinrud H, Hagen KB, Kvien TK. Physiotherapy interventions
(proprioceptive training) for patients with rheumatoid arthritis. for ankylosing spondylitis. Cochrane Database of Systematic
Cochrane Database of Systematic Reviews 2010, Issue 5. [DOI: Reviews 2008, Issue 1. [DOI: 10.1002/14651858.CD002822.pub3]
10.1002/14651858.CD007648.pub2]
Dahm 2010
van der Heijden 2015 Dahm KT, Brurberg KG, Jamtvedt G, Hagen KB. Advice to rest
van der Heijden RA, Lankhorst NE, van Linschoten R, in bed versus advice to stay active for acute low-back pain and
Bierma-Zeinstra SMA, van Middelkoop M. Exercise sciatica. Cochrane Database of Systematic Reviews 2010, Issue 6.
for treating patellofemoral pain syndrome. Cochrane [DOI: 10.1002/14651858.CD007612.pub2]
Database of Systematic Reviews 2015, Issue 1. [DOI:
10.1002/14651858.CD010387.pub2] Dal Bello-Haas 2013
Dal Bello-Haas V, Florence JM. Therapeutic exercise for people
Yamato 2015 with amyotrophic lateral sclerosis or motor neuron disease.
Yamato TP, Maher CG, Saragiotto BT, Hancock MJ, Ostelo RWJG, Cochrane Database of Systematic Reviews 2013, Issue 5. [DOI:
Cabral CMN, et al. Pilates for low back pain. Cochrane 10.1002/14651858.CD005229.pub3]
10.1002/14651858.CD010265.pub2] de Souza 2012
de Souza RF, Lovato da Silva CH, Nasser M, Fedorowicz Z,
Al-Muharraqi MA. Interventions for the management
References to excluded reviews of temporomandibular joint osteoarthritis. Cochrane
Aggarwal 2011 Database of Systematic Reviews 2012, Issue 4. [DOI:
10.1002/14651858.CD007261.pub2]
Aggarwal VR, Lovell K, Peters S, Javidi H, Joughin A,
Goldthorpe J. Psychosocial interventions for the Fokkenrood 2013
management of chronic orofacial pain. Cochrane
Fokkenrood HJP, Bendermacher BLW, Lauret GJ,
Willigendael EM, Prins MH, Teijink JAW. Supervised exercise
10.1002/14651858.CD008456.pub2]
therapy versus non-supervised exercise therapy for intermittent
Bierma-Zeinstra 2011 claudication. Cochrane Database of Systematic Reviews 2013,
Issue 8. [DOI: 10.1002/14651858.CD005263.pub3]
Bierma-Zeinstra SMA, Brinks A, Verhagen AP, Van Rijn RM,
Koes BW, Verhaar JAN. Interventions for lateral hip pain Franke 2015
(tendinopathy or bursitis). Cochrane Database of Systematic
Franke H, Fryer G, Ostelo RWJG, Kamper SJ. Muscle energy
Reviews 2011, Issue 1. [DOI: 10.1002/14651858.CD008924]
technique for non-specific low-back pain. Cochrane
Brønfort 2014 Database of Systematic Reviews 2015, Issue 2. [DOI:
10.1002/14651858.CD009852.pub2]
Brønfort G, Haas M, Evans RL, Goldsmith CH, Assendelft WJJ,
Bouter LM. Non-invasive physical treatments for chronic/ Green 2003
recurrent headache. Cochrane Database of Systematic Reviews
Green S, Buchbinder R, Hetrick SE. Physiotherapy interventions
2014, Issue 8. [DOI: 10.1002/14651858.CD001878.pub3]
for shoulder pain. Cochrane Database of Systematic Reviews
Brønfort 2015 2003, Issue 2. [DOI: 10.1002/14651858.CD004258]
Brønfort G, Evans RL, Goldsmith CH, Haas M, Leininger B, Gross 1998
Levin M, et al. Spinal rehabilitative exercise and manual
Gross A, Aker P, Goldsmith CH, Peloso PMJ. Physical
treatment for the prevention of migraine attacks in adults.
medicine modalities for mechanical neck disorders.
Cochrane Database of Systematic Reviews 2015, Issue 8. [DOI:
10.1002/14651858.CD011848]
10.1002/14651858.CD000961]
Choi 2010
Gross 2012
Choi BKL, Verbeek JH, Tam WWS, Jiang JY. Exercises for
Gross A, Forget M, St George K, Fraser MMH, Graham N,
prevention of recurrences of low-back pain. Cochrane
Perry L, et al. Patient education for neck pain. Cochrane
10.1002/14651858.CD006555.pub2]
10.1002/14651858.CD005106.pub4]
Collaboration.
Informed decisions.

Gross 2015b Jordan 2010

Gross A, Langevin P, Burnie SJ, Bédard-Brochu MS, Empey B, Jordan JL, Holden MA, Mason EEJ, Foster NE. Interventions to
Dugas E, et al. Manipulation and mobilisation for neck pain improve adherence to exercise for chronic musculoskeletal pain
contrasted against an inactive control or another active in adults. Cochrane Database of Systematic Reviews 2010, Issue
treatment. Cochrane Database of Systematic Reviews 2015, Issue 1. [DOI: 10.1002/14651858.CD005956.pub2]
9. [DOI: 10.1002/14651858.CD004249.pub4]
Kamper 2014
Hayden 2012 Kamper SJ, Apeldoorn AT, Chiarotto A, Smeets RJ, Ostelo RWJG,
Hayden JA, Cartwright J, van Tulder MW, Malmivaara A. Guzman J, et al. Multidisciplinary biopsychosocial rehabilitation
Exercise therapy for chronic low back pain. Cochrane for chronic low back pain. Cochrane Database of Systematic
Database of Systematic Reviews 2012, Issue 4. [DOI: Reviews 2014, Issue 9. [DOI: 10.1002/14651858.CD000963.pub3]
10.1002/14651858.CD009790]
Karjalainen 1999
Heintjes 2003 Karjalainen KA, Malmivaara A, van Tulder MW, Roine R,
Heintjes EM, Berger M, Bierma-Zeinstra SMA, Bernsen RMD, Jauhiainen M, Hurri H, et al. Multidisciplinary rehabilitation for
Verhaar JAN, Koes BW. Exercise therapy for patellofemoral pain fibromyalgia and musculoskeletal pain in working age adults.
syndrome. Cochrane Database of Systematic Reviews 2003, Issue Cochrane Database of Systematic Reviews 1999, Issue 3. [DOI:
4. [DOI: 10.1002/14651858.CD003472] 10.1002/14651858.CD001984]
Henschke 2010 Karjalainen 2003

Henschke N, Ostelo RWJG, van Tulder MW, Vlaeyen JWS, Karjalainen KA, Malmivaara A, van Tulder MW, Roine R,
Morley S, Assendelft WJJ, et al. Behavioural treatment for Jauhiainen M, Hurri H, et al. Multidisciplinary biopsychosocial
chronic low-back pain. Cochrane Database of Systematic rehabilitation for neck and shoulder pain among working age
Reviews 2010, Issue 7. [DOI: 10.1002/14651858.CD002014.pub3] adults. Cochrane Database of Systematic Reviews 2003, Issue 2.
[DOI: 10.1002/14651858.CD002194]
Heymans 2004
Heymans MW, van Tulder MW, Esmail R, Bombardier C, Larun 2016
Koes BW. Back schools for non-specific low-back pain. Larun L, Brurberg KG, Odgaard-Jensen J, Price JR.
Cochrane Database of Systematic Reviews 2004, Issue 4. [DOI: Exercise therapy for chronic fatigue syndrome. Cochrane
10.1002/14651858.CD000261.pub2] Database of Systematic Reviews 2016, Issue 2. [DOI:
10.1002/14651858.CD003200.pub4]
Hilde 2006
Hilde G, Hagen KB, Jamtvedt G, Winnem M. Advice to stay Liddle 2015
active as a single treatment for low-back pain and sciatica. Liddle SD, Pennick V. Interventions for preventing and
Cochrane Database of Systematic Reviews 2006, Issue 2. [DOI: treating low-back and pelvic pain during pregnancy.
10.1002/14651858.CD003632.pub2] Cochrane Database of Systematic Reviews 2015, Issue 9. [DOI:
10.1002/14651858.CD001139.pub4]
Hoving 2014
Hoving JL, Lacaille D, Urquhart DM, Hannu TJ, Sluiter JK, Liu 2013
Frings-Dresen MHW. Non-pharmacological interventions for Liu H, Yao K, Zhang J, Li L, Wu T, Brox JI, et al. Sling
preventing job loss in workers with inflammatory arthritis. exercise therapy for chronic low-back pain. Cochrane
Cochrane Database of Systematic Reviews 2014, Issue 11. [DOI: Database of Systematic Reviews 2013, Issue 9. [DOI:
10.1002/14651858.CD010208.pub2] 10.1002/14651858.CD010689]
Hurley 2013 Miller 2014

Hurley M, Dickson K, Walsh N, Hauari H, Grant R, Cumming J, Miller J, Gross A, Kay TM, Graham N, Burnie SJ, Goldsmith CH,
et al. Exercise interventions and patient beliefs for people et al. Manual therapy with exercise for neck pain. Cochrane
with chronic hip and knee pain: a mixed methods review. Database of Systematic Reviews 2014, Issue 7. [DOI:
Cochrane Database of Systematic Reviews 2013, Issue 12. [DOI: 10.1002/14651858.CD011225]
10.1002/14651858.CD010842]
Moi 2013
IJzelenberg 2011 Moi JHY, Sriranganathan MK, Edwards CJ, Buchbinder R.
IJzelenberg W, Rubinstein SM, Hayden J, Koes BW, Lifestyle interventions for chronic gout. Cochrane
van Tulder MW. Exercise therapy for acute non-specific low-back Database of Systematic Reviews 2013, Issue 5. [DOI:
pain. Cochrane Database of Systematic Reviews 2011, Issue 10. 10.1002/14651858.CD010039.pub2]
[DOI: 10.1002/14651858.CD009365]
O'Brien 2004
Jones 2000 O'Brien K, Nixon S, Glazier R, Tynan AM. Progressive resistive
Jones G, Crotty M, Brooks P. Interventions for treating psoriatic exercise interventions for adults living with HIV/AIDS.
arthritis. Cochrane Database of Systematic Reviews 2000, Issue Cochrane Database of Systematic Reviews 2004, Issue 4. [DOI:
3. [DOI: 10.1002/14651858.CD000212] 10.1002/14651858.CD004248.pub2]
Collaboration.
Informed decisions.

O'Connell 2013 Database of Systematic Reviews 2012, Issue 1. [DOI:

O'Connell NE, Wand BM, McAuley J, Marston L, Moseley GL. 10.1002/14651858.CD008921.pub2]
Interventions for treating pain and disability in adults with
Riemsma 2003
complex regional pain syndrome- an overview of systematic
reviews. Cochrane Database of Systematic Reviews 2013, Issue 4. Riemsma RP, Kirwan JR, Taal E, Rasker HJJ. Patient
[DOI: 10.1002/14651858.CD009416.pub2] education for adults with rheumatoid arthritis. Cochrane
Page 2012 10.1002/14651858.CD003688]
Page MJ, O'Connor D, Pitt V, Massy-Westropp N. Exercise
Schaafsma 2013
and mobilisation interventions for carpal tunnel syndrome.
Cochrane Database of Systematic Reviews 2012, Issue 6. [DOI: Schaafsma FG, Whelan K, van der Beek AJ, van der Es-
10.1002/14651858.CD009899] Lambeek LC, Ojajärvi A, Verbeek JH. Physical conditioning as
part of a return to work strategy to reduce sickness absence
Page 2014 for workers with back pain. Cochrane Database of Systematic
Page MJ, Green S, Kramer S, Johnston RV, McBain B, Chau M, et Reviews 2013, Issue 8. [DOI: 10.1002/14651858.CD001822.pub3]
al. Manual therapy and exercise for adhesive capsulitis (frozen
Steultjens 2004
shoulder). Cochrane Database of Systematic Reviews 2014, Issue
8. [DOI: 10.1002/14651858.CD011275] Steultjens EEMJ, Dekker JJ, Bouter LM, Schaardenburg DD,
Kuyk MAMAH, Van den Ende ECHM. Occupational therapy for
Peters 2013 rheumatoid arthritis. Cochrane Database of Systematic Reviews
Peters S, Page MJ, Coppieters MW, Ross M, Johnston V. 2004, Issue 1. [DOI: 10.1002/14651858.CD003114.pub2]
Rehabilitation following carpal tunnel release. Cochrane
Stones 2005
10.1002/14651858.CD004158.pub2] Stones W, Cheong YC, Howard FM, Singh S. Interventions
for treating chronic pelvic pain in women. Cochrane
Preston 2004 Database of Systematic Reviews 2005, Issue 2. [DOI:
Preston NJ, Seers K, Mortimer PS. Physical therapies for 10.1002/14651858.CD000387]
reducing and controlling lymphoedema of the limbs.
Takken 2008
10.1002/14651858.CD003141.pub2] Takken T, Van Brussel M, Engelbert RH, van der Net JJ, Kuis W,
Helders PPJM. Exercise therapy in juvenile idiopathic arthritis.
Proctor 2007 Cochrane Database of Systematic Reviews 2008, Issue 2. [DOI:
Proctor M, Murphy PA, Pattison HM, Suckling JA, Farquhar C. 10.1002/14651858.CD005954.pub2]
Behavioural interventions for dysmenorrhoea. Cochrane
van der Heijden 2013
10.1002/14651858.CD002248.pub3] van der Heijden RA, Lankhorst NE, van Linschoten R,
Bierma-Zeinstra SMA, van Middelkoop M. Exercise
Radner 2012 for treating patellofemoral pain syndrome. Cochrane
Radner H, Ramiro S, Buchbinder R, Landewé RBM, van Database of Systematic Reviews 2013, Issue 2. [DOI:
der Heijde D, Aletaha D. Pain management for inflammatory 10.1002/14651858.CD010387]
arthritis (rheumatoid arthritis, psoriatic arthritis, ankylosing
van Dessel 2014
spondylitis and other spondyloarthritis) and gastrointestinal
or liver comorbidity. Cochrane Database of Systematic Reviews van Dessel N, den Boeft M, van der Wouden JC, Kleinstäuber M,
2012, Issue 1. [DOI: 10.1002/14651858.CD008951.pub2] Leone SS, Terluin B, et al. Non-pharmacological
interventions for somatoform disorders and medically
Regnaux 2012 unexplained physical symptoms (MUPS) in adults. Cochrane
Regnaux JP, Trinquart L, Boutron I, Nguyen C, Brosseau L, Database of Systematic Reviews 2014, Issue 11. [DOI:
Ravaud P. High-intensity versus low-intensity physical 10.1002/14651858.CD011142.pub2]
activity or exercise in patients with hip or knee osteoarthritis.
White 2004
10.1002/14651858.CD010203] White CM, Pritchard J, Turner-Stokes L. Exercise for people
with peripheral neuropathy. Cochrane Database of Systematic
Regnaux 2014 Reviews 2004, Issue 4. [DOI: 10.1002/14651858.CD003904.pub2]
Regnaux JP, Lefevre-Colau MM, Palazzo C, Poiraudeau S,
Zammit 2010
Ravaud P, Boutron I. Exercise therapy for ankylosing spondylitis.
Cochrane Database of Systematic Reviews 2014, Issue 10. [DOI: Zammit GV, Menz HB, Munteanu SE, Landorf KB, Gilheany MF.
10.1002/14651858.CD011321] Interventions for treating osteoarthritis of the big toe joint.
Richards 2012 10.1002/14651858.CD007809.pub2]
Richards BL, Whittle SL, Buchbinder R. Neuromodulators
for pain management in rheumatoid arthritis. Cochrane
Collaboration.
Informed decisions.

Østerås 2013 Collins 1997

Østerås N, Kjeken I, Smedslund G, Moe RH, Slatkowsky- Collins SL, Moore RA, McQuay HJ. The visual analogue pain
Christensen B, Uhlig T, et al. Exercises for hand osteoarthritis. intensity scale: what is moderate pain in millimetres?. Pain
Cochrane Database of Systematic Reviews 2013, Issue 2. [DOI: 1997;72(1-2):95-7. [DOI: 10.1016/S0304-3959(97)00005-5;
10.1002/14651858.CD010388] PUBMED: 9272792]
Daenen 2015
Additional references Daenen L, Varkey E, Kellmann M, Nijs J. Exercise, not to exercise,
Abdulla 2013 or how to exercise in patients with chronic pain? Applying
science to practice. Clinical Journal of Pain 2015;31(2):108-14.
Abdulla A, Adams N, Bone M, Elliott AM, Gaffin J, Jones D, et al.
[DOI: 10.1097/AJP.0000000000000099; PUBMED: 24662498]
British Geriatric Society. Guidance on the management of pain
in older people. Age and Ageing 2013;42(Suppl 1):i1-57. [DOI: Deschartes 2013
10.1093/ageing/afs200; PUBMED: 23420266]
Deschartes A, Trinquart L, Boutron I, Ravaud P. Influence
American College of Sports Medicine 2007 of trial sample size on treatment effect estimates: meta-
epidemiological study. BMJ 2013;346:f2304. [DOI: 10.1136/
American College of Sports Medicine, American Heart
bmj.f2304]
Association. Exercise and acute cardiovascular events:
placing the risks into perspective. Medicine and Science Dworkin 2005
in Sport and Exercise 2007;39(5):886-97. [DOI: 10.1249/
Dworkin RH, Turk DC, Farrar JT, Haythornthwaite JA, Jensen MP,
mss.0b013e3180574e0e; PUBMED: 17468590]
Katz NP, et al. Core outcome measures for chronic pain clinical
Benton 2011 trials: IMMPACT recommendations. Pain 2005;113:9-19.
[PUBMED: 15621359]
Benton MJ, Kasper MJ, Raab SA, Waggener GT, Swan PD.
Short-term effects of resistance training frequency on body Dworkin 2008
composition and strength in middle-aged women. Journal
Dworkin RH, Turk DC, Wyrwich KW, Beaton D, Cleeland CS,
of Strength and Conditioning Research/National Strength &
Farrar JT, et al. Consensus Statement: Interpreting the
Conditioning Association 2011;25(11):3142-9. [DOI: 10.1519/
clinical importance of treatment outcomes in chronic pain
JSC.0b013e31820f505f]
clinical trials: IMMPACT recommendations. Journal of Pain
Bigal 2012 2008;9(2):105-21. [DOI: 10.1016/j.jpain.2007.09.005; PUBMED:
18055266]
Bigal ME, Rapoport AM. Obesity and chronic daily headache.
Current Pain and Headache Reports 2012;16(1):101-9. [DOI: Eccleston 2014
10.1007/s11916-011-0232-0; PUBMED: 22076673]
Eccleston C, Fisher E, Craig L, Duggan GB, Rosser BA,
Bouchard 2012 Keogh E. Psychological therapies (Internet-delivered)
for the management of chronic pain in adults. Cochrane
Bouchard C, Blair SN, Church TS, Earnest CP, Hagberg JM,
Häkkinen K, et al. Adverse metabolic response to regular
10.1002/14651858.CD010152.pub2]
exercise: is it a rare or common occurrence?. PLoS One
2012;7(5):e37887. [DOI: 10.1371/journal.pone.0037887; Elliott 1999
PUBMED: 22666405]
Elliott AM, Smith BH, Penny KI, Smith WC, Chambers WA.
Bousema 2007 The epidemiology of chronic pain in the community. Lancet
1999;354:1248-52. [PUBMED: 10520633]
Bousema EJ, Verbunt JA, Seelen HA, Vlaeyen JW, Knottnerus JA.
Disuse and physical deconditioning in the first year after Elzahaf 2012
the onset of back pain. Pain 2007;130(3):279-86. [PUBMED:
Elzahaf RA, Tashani OA, Unsworth BA, Johnson MI. The
17467902]
prevalence of chronic pain with an analysis of countries with a
Breivik 2006 Human Development Index less than 0.9: a systematic review
without meta-analysis. Current Medical Research & Opinion
Breivik H, Collett B, Ventafridda V, Cohen R, Gallacher D. Survey
2012;28(7):1221-9. [DOI: 10.1185/03007995.2012.703132]
of chronic pain in Europe: prevalence, impact on daily life,
and treatment. European Journal of Pain 2006;10(4):287-333. Engers 2008
[PUBMED: 16095934]
Engers A, Jellema P, Wensing M, van der Windt DA, Grol R,
Cohen 1988 van Tulder MW. Individual patient education for low back pain.
Cohen J. Statistical Power Analysis for the Behavioral Sciences.
10.1002/14651858.CD004057.pub3; PUBMED: 18254037]
2nd Edition. Hillsdale, NJ: L. Erlbaum Associates, 1988. [ISBN
0805802835] Enoka 1997
Enoka RM. Neural adaptations with chronic physical activity.
Journal of Biomechanics 1997;30(5):447-55. [PUBMED: 9109556]
Collaboration.
Informed decisions.

Exercise is Medicine Messier 2013

American College of Sports Medicine (ACSM). Exercise is Messier SP, Mihalko SL, Legault C, Miller GD, Nicklas BJ, DeVita P,
Medicine. http://www.exerciseismedicine.org/ (accessed 27 et al. Effects of intensive diet and exercise on knee joint loads,
June 2014). inflammation, and clinical outcomes among overweight and
obese adults with knee osteoarthritis: the IDEA randomized
Finan 2013 clinical trial. JAMA 2013;310(12):1263-73. [DOI: 10.1001/
Finan PH, Smith MT. The comorbidity of insomnia, chronic jama.2013.277669; PUBMED: 24065013]
pain, and depression: dopamine as a putative mechanism.
Sleep Medicine Reviews 2013;17(3):173-83. [DOI: 10.1016/ Mooney 2006
j.smrv.2012.03.003; PUBMED: 22748562] Mooney V, Verna J, Morris C. Clinical management of
chronic, disabling low back syndromes. In: Morris C editor(s).
Gross 2009 Rehabilitation of the Spine: a Practitioner's Manual. New York:
Gross A, Forget M, St George K, Fraser MM, Graham N, McGraw-Hill, 2006.
Perry L, et al. Patient education for neck pain. Cochrane
Database of Systematic Reviews 2012, Issue 3. [DOI: Moore 1998
10.1002/14651858.CD005106.pub4; PUBMED: 22419306] Moore RA, Gavaghanb D, Tramer MR, Collins SL, McQuay HJ.
Size is everything - large amounts of information are needed
Harvard 2013 to overcome random effects in estimating direction and
Better balance: mental and physical fitness are both essential: magnitude of treatment effects. Pain 1998;78:209-16. [PUBMED:
daily physical activity, targeted balance exercise, and steps to 9870574]
maintain your brain all help prevent falls. Harvard Men's Health
Watch 2013;17(11):5. [PUBMED: 24000423] Moore 2008
Moore RA, Barden J, Derry S, McQuay HJ. Managing potential
Hoffman 2014 publication bias. In: McQuay HJ, Kalso E, Moore RA editor(s).
Hoffmann TC, Glasziou PP, Boutron I, Milne R, Perera R, Systematic Reviews in Pain Research: Methodology Refined.
Moher D, et al. Better reporting of interventions: template for Seattle: IASP Press, 2008:15-23. [ISBN: 978-0-931092-69-5]
intervention description and replication (TIDieR) checklist
and guide. BMJ 2014;348. [DOI: 10.1136/bmj.g1687; PUBMED: Moore 2010
24609605] Moore RA, Moore OA, Derry S, Peloso PM, Gammaitono AR,
Wang H. Responder analysis for pain relief and numbers needed
IDF 2012 to treat in a meta-analysis of etoricoxib osteoarthritis trials:
Guariguata L. By the numbers: new estimates from the IDF bridging a gap between clinical trials and clinical practice.
Diabetes Atlas Update for 2012. Diabetes Research and Clinical Annals of Rheumatic Diseases 2010;69(2):374-9. [DOI: 10.1136/
Practice 2012;98(3):524-5. [DOI: 10.1016/j.diabres.2012.11.006; ard.2009.107805; PUBMED: 19364730]
PUBMED: 23217268]
Moore 2013
IntHout 2015 Moore RA, Straube S, Aldington D. Pain measures and cut-offs
IntHout J, Ioannidis JPA, Borm GF, Goeman JJ. Small studies - 'no worse than mild pain' as a simple, universal outcome.
are more heterogeneous than large ones: a meta-meta-analysis. Anaesthesia 2013;68(4):400-12. [DOI: 10.1111/anae.12148;
Journal of Clinical Epidemiology 2015;68(8):860-9. [DOI: 10.1016/ PUBMED: 23347230]
j.jclinepi.2015.03.017; PUBMED: 25959635]
Moore 2014a
Leadley 2014 Moore RA, Derry S, Taylor RS, Straube S, Phillips CJ. The
Leadley RM, Armstrong N, Reid KJ, Allen A, Misso KV, Kleijnen J. costs and consequences of adequately managed chronic
Healthy aging in relation to chronic pain and quality of life non-cancer pain and chronic neuropathic pain. Pain Practice
in Europe. Pain Practice 2014;14(6):547-58. [DOI: 10.1111/ 2014;14(1):79-94. [PUBMED: 23464879]
papr.12125; PUBMED: 24138082]
Moore 2014b
Mayer 2008 Moore RA, Derry S, Taylor RS, Straube S, Phillips CJ. The costs
Mayer J, Mooney V, Dagenais S. Evidence-informed and consequences of adequately managed chronic non-cancer
management of low back pain with lumbar extensor pain and chronic neuropathic pain. Pain Practice 2014; Vol. 14,
strengthening exercises. The Spine Journal 2008;8(1):96-113. issue 1:79-94. [DOI: 10.1111/papr.12050]
[DOI: 10.1016/j.spinee.2007.09.008; PUBMED: 18164458]
National Library of Medicine
Merskey 2011 National Library of Medicine. Medical Subject Headings. search
Merskey H, Bogduk N. Part III: Pain terms: a current list with "chronic pain" at www.nlm.nih.gov/mesh/MBrowser.html
definitions and notes on usage. Classification of Chronic Pain. (accessed 22 August 2014). [MeSH unique ID: D059350]
2nd Edition. Seattle: IASP Task Force on Taxonomy, IASP Press,
2011 [update from 1994]. NICE 2014
National Institute for Health and Care Excellence (NICE).
Osteoarthritis. Care and management in adults (CG177),
Collaboration.
Informed decisions.

February 2014. www.nice.org.uk/guidance/cg177 (accessed 4 SIGN 2013

December 2015). [ISBN: 978-1-4731-0426-6] Scottish Intercollegiate Guidelines Network (SIGN).
Management of chronic pain. SIGN publication no. 136. 1st
Nnoaham 2008
Edition. Edinburgh, UK: SIGN, 2013. [ISBN: 978-1-909103-17-7]
Nnoaham KE, Kumbang J. Transcutaneous electrical
nerve stimulation (TENS) for chronic pain. Cochrane Smith 2007
Database of Systematic Reviews 2008, Issue 3. [DOI: Smith BH, Macfarlane GJ, Torrance N. Epidemiology of
10.1002/14651858.CD003222.pub2; PUBMED: 18646088] chronic pain, from the laboratory to the bus stop: time to add
understanding of biological mechanisms to the study of risk
Nüesch 2010
factors in population-based research?. Pain 2007;127(1-2):5-10.
Nüesch E, Trelle S, Reichenbach S, Rutjes AW, Tschannen B, [DOI: 10.1016/j.pain.2006.11.001; PUBMED: 17140732]
Altman DG, et al. Small study effects in meta-analyses
of osteoarthritis trials: meta-epidemiological study. BMJ Stagg 2011
2010;341:c3515. [PUBMED: 20639294] Stagg NJ, Mata HP, Ibrahim MM, Henriksen EJ, Porreca F,
Vanderah TW, et al. Regular exercise reverses sensory
Nüesch 2011
hypersensitivity in a rat neuropathic pain model: role of
Nüesch E, Dieppe P, Reichenbach S, Williams S, Iff S, Jüni P. endogenous opioids. Anaesthesiology 2011;114(4):940-8. [DOI:
All cause and disease specific mortality in patients with knee 10.1097/ALN.0b013e318210f880; PUBMED: 21386701]
or hip osteoarthritis: population based cohort study. BMJ
2011;342:d1165. [10.1136/bmj.d1165] Straube 2015
Straube S, Derry S, Moore RA, McQuay HJ. Vitamin D for
Park 2012
the treatment of chronic painful conditions in adults.
Park J, Hughes AK. Non-pharmacological approaches to Cochrane Database of Systematic Reviews 2015, Issue 5. [DOI:
the management of chronic pain in community-dwelling 10.1002/14651858.CD007771.pub3; PUBMED: 25946084]
older adults: a review of empirical evidence. Journal of the
American Geriatric Society 2012;60(3):555-68. [DOI: 10.1111/ Thornton 2000
j.1532-5415.2011.03846.x; PUBMED: 22288789] Thornton A, Lee P. Publication bias in meta-analysis: its
causes and consequences. Journal of Clinical Epidemiology
Perruchoud 2014
2000;53(2):207-16. [DOI: 10.1016/S0895-4356(99)00161-4]
Perruchoud C, Buchser E, Johanek LM, Aminian K, Paraschiv-
Ionescu A, Taylor RS. Assessment of physical activity of patients Timmons 2014
with chronic pain. Neuromodulation 2014;17:42-7. [DOI: Timmon J, Phillips SM. Do we have a valid public health
10.1111/ner.12036] strategy for exercise prescription?. Sport and Exercise Scientist
2014;39:26-7.
RevMan 2014 [Computer program]
The Nordic Cochrane Centre, Cochrane. Review Manager To 2012
(RevMan). Version 5.3. Copenhagen: The Nordic Cochrane To T, Stanojevic S, Moores G, Gershon AS, Bateman ED, Cruz AA,
Centre, Cochrane, 2014. et al. Global asthma prevalence in adults: findings from
the cross-sectional world health study. BMC Public Health
Sakuma 2012
2012;12:204. [DOI: 10.1186/1471-2458-12-204; PUBMED:
Sakuma M, Endo N. Space flight/bedrest immobilization and 22429515]
bone. Exercise training for osteoporosis. Clinical Calcium
2012;22(12):1903-7. [DOI: CliCa121219031907; PUBMED: Torrance 2010
23187084] Torrance N, Elliott AM, Lee AJ, Smith BH. Severe chronic pain
is associated with increased 10 year mortality. A cohort record
Schofield 2011
linkage study. European Journal of Pain 2010;14(4):380-6. [DOI:
Schofield P, Clarke A, Jones D, Martin D, McNamee P, Smith BH. 10.1016/j.ejpain.2009.07.006]
Chronic pain in later life: a review of current issues and
challenges. Aging Health 2011;7(4):551-6. [DOI: 10.2217/ Turk 2003
ahe.11.41] Turk DC, Dworkin RH, Allen RR, Bellamy N, Brandenburg N,
Carr DB, et al. Core outcome domains for chronic pain clinical
Shea 2007
trials: IMMPACT recommendations. Pain 2003;106:337-45.
Shea BJ, Grimshaw JM, Wells GA, Boers M, Andersson N, [PUBMED: 14659516]
Hamel C, et al. Development of AMSTAR: a measurement
tool to assess the methodological quality of systematic Turner 2013
reviews. BMC Medical Research Methodology 2007;7:10. [DOI: Turner RM, Bird SM, Higgins JPT. The impact of study size
10.1186/1471-2288-7-10; PUBMED: 17302989] on meta-analyses: examination of underpowered studies in
Cochrane Reviews. PLoS One 2013;8(3):e59202. [DOI: 10.1371/
journal.pone.0059202; PUBMED: 23544056]
Collaboration.
Informed decisions.

Ursini 2011 Waddell 1987

Ursini F, Naty S, Grembiale RD. Fibromyalgia and obesity: the Waddell G. 1987 Volvo award in clinical sciences. A new
hidden link. Rheumatology International 2011;31(11):1403-8. clinical model for the treatment of low-back pain. Spine
[DOI: 10.1007/s00296-011-1885-z; PUBMED: 21476098] 1987;12(7):632-44. [PUBMED: 2961080]
van Hecke 2013a WHO 2015

van Hecke O, Torrance N, Smith BH. Chronic pain epidemiology World Health Organization. Physical activity. Fact sheet 385.
- where do lifestyle factors fit in?. British Journal of Pain 2013; www.who.int/mediacentre/factsheets/fs385/en/ (accessed 4
Vol. 7, issue 4:209-17. [DOI: 10.1177/2049463713493264] December 2015).
van Hecke 2013b Wieland 2013

van Hecke O, Torrance N, Smith BH. Chronic pain epidemiology Wieland LS, Skoetz N, Manheimer E, Pilkington K, Vempati R,
and its clinical relevance. British Journal of Anaesthesia Berman BM. Yoga treatment for chronic non-specific low-back
2013;111(1):13-8. [DOI: 10.1093/bja/aet123; PUBMED: pain. Cochrane Database of Systematic Reviews 2013, Issue 7.
23794640] [DOI: 10.1002/14651858.CD010671]
Vigorito 2014 Williams 2012

Vigorito C, Giallauria F. Effects of exercise on cardiovascular Williams ACDC, Eccleston C, Morley S. Psychological therapies
performance in the elderly. Frontiers in Physiology 2014;5:51. for the management of chronic pain (excluding headache) in
[DOI: 10.3389/fphys.2014.00051] adults. Cochrane Database of Systematic Reviews 2012, Issue 11.
[DOI: 10.1002/14651858.CD007407.pub3]
Vos 2015
Vos T, Barber RM, Bell B, Bertozzi-Villa A, Biryukov S, Bolliger I,
et al. Global Burden of Disease Study 2013 Collaborators. References to other published versions of this review
Global, regional, and national incidence, prevalence, and Geneen 2014
years lived with disability for 301 acute and chronic diseases
Geneen L, Smith B, Clarke C, Martin D, Colvin LA, Moore RA.
and injuries in 188 countries, 1990-2013: a systematic
Physical activity and exercise for chronic pain in adults: an
analysis for the Global Burden of Disease Study 2013. Lancet
overview of Cochrane reviews. Cochrane Database of Systematic
2015;386(9995):743-800. [DOI: 10.1016/S0140-6736(15)60692-4;
Reviews 2014, Issue 8. [DOI: 10.1002/14651858.CD011279]
PUBMED: 26063472]

ADDITIONAL TABLES

Table 1. AMSTAR tool to assess the methodological quality of systematic reviews
Criteria Specific requirements (possible answers: yes, no, cannot answer, not applicable)
1. Was an 'a priori' design The research question and inclusion criteria should be established before the conduct of the re-
used? view.
Note: need to refer to a protocol, ethics approval, or predetermined/a priori published research objec-
tives to score a "yes."
2. Was there duplicate study There should be at least 2 independent data extractors and a consensus procedure for disagree-
selection and data extraction? ments should be in place.
Note: 2 people do study selection, 2 people do data extraction, consensus process or 1 person checks
the other person's work.
3. Was a comprehensive litera- At least 2 electronic sources should be searched. The report must include years and databases
ture search performed? used (e.g. CENTRAL, MEDLINE, and Embase). Keywords or MeSH terms (or both) must be stated and
where feasible the search strategy should be provided. All searches should be supplemented by
consulting current contents, reviews, textbooks, specialised registers, or experts in the particular
field of study, and by reviewing the references in the studies found.
Note: if at least 2 sources + 1 supplementary strategy used, select "yes"(Cochrane register/ CENTRAL
counts as 2 sources; a grey literature search counts as supplementary).
4. Was the status of the pub- The authors should state that they searched for reports regardless of their publication type. The
lication (i.e. grey literature) authors should state whether or not they excluded any reports (from the systematic review), based
used as inclusion criteria? on their publication status, language, etc.
Collaboration.
Informed decisions.

Table 1. AMSTAR tool to assess the methodological quality of systematic reviews (Continued)

Note: if review indicates that there was a search for "grey literature"or "unpublished literature,"indi-
cate "yes."SIGLE database, dissertations, conference proceedings, and trial registries are all consid-
ered grey for this purpose. If searching a source that contains both grey and non-grey, must specify
that they were searching for grey/unpublished literature.
5. Was a list of studies (includ- A list of included and excluded studies should be provided.
ed and excluded) provided?
Note: acceptable if the excluded studies were referenced. If there was an electronic link to the list but
the link is no longer active, select "no."
6. Were the characteristics of In an aggregated form such as a table, data from the original studies should be provided on the
the included studies provided? participants, interventions, and outcomes. The ranges of characteristics in all the studies analysed,
e.g. age, race, sex, relevant socioeconomic data, disease status, duration, severity, or other dis-
eases should be reported.
Note: acceptable if not in table format as long as they are described as above.
7. Was the scientific quality of 'A priori' methods of assessment should be provided (e.g. for effectiveness studies if the author(s)
the included studies assessed chose to include only randomised, double-blind, placebo-controlled studies, or allocation conceal-
and documented? ment as inclusion criteria); for other types of studies alternative items will be relevant.
Note: can include use of a quality scoring tool or checklist, e.g. Jadad scale, risk of bias, sensitivi-
ty analysis, etc., or a description of quality items, with some type of result for EACH study ("low"or
"high"is acceptable, as long as it is clear which studies scored "low"and which scored "high;"a sum-
mary score/range for all studies is not acceptable).
8. Was the scientific quality of The results of the methodological rigor and scientific quality should be considered in the analysis
the included studies used ap- and the conclusions of the review, and explicitly stated in formulating recommendations.
propriately in formulating con-
clusions? Note: might say something such as "the results should be interpreted with caution due to poor quality
of included studies."Cannot score "yes"for this question if scored "no"for question 7.
9. Were the methods used to For the pooled results, a test should be done to ensure the studies were combinable, to assess their
combine findings of studies homogeneity (i.e. Chi2 test for homogeneity, I2 statistic). If heterogeneity exists, a random-effects
appropriate? model should be used or the clinical appropriateness of combining should be taken into considera-
tion (i.e. is it sensible to combine?), or both.
Note: indicate "yes"if they mention or describe heterogeneity, i.e. if they explain that they cannot pool
because of heterogeneity/variability between interventions.
10. Was the likelihood of publi- An assessment of publication bias should include a combination of graphical aids (e.g. funnel plot,
cation bias assessed? other available tests) or statistical tests (e.g. Egger regression test), or both.
Note: if no test values or funnel plot included, score "no."Score "yes"if they mention that publication
bias could not be assessed because there were fewer than 10 included studies.
11. Was the conflict of interest Potential sources of support should be clearly acknowledged in both the systematic review and the
stated? included studies.
Note: to get a "yes,"must indicate source of funding or support for the systematic review AND for each
of the included studies.

Table 2. Reasons for exclusion
Review Reason for exclusion from overview
Aggarwal 2011 Not exercise/physical activity
Collaboration.
Informed decisions.

Table 2. Reasons for exclusion (Continued)

Brønfort 2015 Protocol stage only - possibly include when published as full review
Bierma-Zeinstra 2011 Protocol stage only - exclude when published as full review
Brønfort 2014 Withdrawn from the Cochrane Library
Choi 2010 Not chronic using definition of > 3 months
Craane 2006 Protocol stage only - possibly include when published as full review
Dagfinrud 2008 Physiotherapy - required therapist to perform intervention
Dahm 2010 Acute pain, not chronic. Intervention was advice
Dal Bello-Haas 2013 Malignant condition
de Souza 2012 Drug- and surgery-based interventions
Fokkenrood 2013 Did not include RCTs (excluded studies with control groups)
Franke 2015 Not exercise/physical activity
Green 2003 Physiotherapy - required therapist to perform intervention
Gross 1998 Withdrawn from the Cochrane Library
Gross 2012 Not exercise/physical activity
Gross 2015b Not exercise/physical activity
Hayden 2012 Protocol stage only - possibly include when published as full review
Heintjes 2003 Withdrawn from the Cochrane Library January 2015
Henschke 2010 Not exercise/physical activity
Heymans 2004 Exercise could not be assessed as stand-alone intervention
Hilde 2006 Withdrawn from the Cochrane Library
Hoving 2014 No exercise intervention, and no pain outcome measure
Hurley 2013 Protocol stage only - exclude when published as full review
IJzelenberg 2011 Protocol stage only - exclude when published as full review
Jones 2000 Drug-based interventions
Jordan 2010 Intervention to improve adherence to exercise, not exercise itself
Kamper 2014 Exercise could not be assessed as stand-alone intervention
Karjalainen 1999 Exercise could not be assessed as stand-alone intervention
Karjalainen 2003 Exercise could not be assessed as stand-alone intervention
Collaboration.
Informed decisions.

Table 2. Reasons for exclusion (Continued)

Larun 2016 Chronic fatigue, not chronic pain
Liddle 2015 Pain in pregnancy only, not chronic pain
Liu 2013 Protocol stage only - unsure about inclusion when published as full review
Miller 2014 Protocol stage only - exclude when published as full review
Moi 2013 Exercise could not be assessed as stand-alone intervention
O'Brien 2004 No pain outcome measure
O'Connell 2013 Overview of reviews, not systematic review
Østerås 2013 Protocol stage only - possibly include when published as full review
Page 2012 No pain outcome measure
Page 2014 Manual therapy - required therapist to perform intervention
Peters 2013 Exercise could not be assessed as stand-alone intervention
Preston 2004 No pain outcome measure
Proctor 2007 Exercise could not be assessed as stand-alone intervention
Radner 2012 Drug-based interventions
Regnaux 2014 Protocol stage only - possibly include when published as full review
Richards 2012 Not exercise/physical activity
Riemsma 2003 Not exercise/physical activity
Schaafsma 2013 No pain outcome measure
Steultjens 2004 Occupational therapy - exercise could not be assessed as stand-alone intervention
Stones 2005 Exercise cannot be assessed as stand-alone intervention
Takken 2008 Aged < 18 years - not adults
van Dessel 2014 Not chronic pain and no specific pain outcome measure
White 2004 No pain outcome measure
Williams 2012 Not exercise/physical activity
Zammit 2010 Surgery or required therapist to perform intervention
RCT: randomised controlled trial.

Collaboration.
Collaboration.
Table 3. Characteristics of included reviews
Review and Assessed as Chronic Duration of Intervention description Control description Outcomes with data re- Time points re-
Cochrane up to date pain condi- pain/ diag- ported ported
Library
Cochrane
Review tion nosis
Group
Bartels 2007 Aug 2007 Hip or knee Not report- All types of exercises devel- No treatment or other Function, quality of life, Post-interven-
OA ed oped in the therapeutic/heat- treatment. mental health, pain, ad- tion (immedi-
Cochrane ed indoor pool (ROM, dynam- verse events ate), 6-month
Muscu-
Better health.
Informed decisions.
Trusted evidence.
ics, aerobics, etc.) were per- follow-up
loskeletal mitted.
Group
Bidonde Oct 2013 Fibromyal- 12 yr (range Aquatic exercise training in- Treatment as usual, Multi-dimensional function Post-interven-
2014 gia 6 to 24) tervention defined as "exer- physical activity as usu- (wellness), self-reported tion (4 to 32 wk)
cise conducted in a vertical al, wait list control, physical function (wellness),
Cochrane standing position." placebo or sham, educa-
Muscu- tion-only, water immer- pain (symptoms),
loskeletal sion-only, and attention
Group stiffness (symptoms),
only.
muscle strength (physical
fitness),
submaximal cardiorespira-
tory function (physical fit-
ness),
withdrawals (safety and ac-

ceptability),
adverse effects (safety and

acceptability)
Boldt 2014 Mar 2011 Spinal cord Mean 66 "Exercise": stretching and Wait list control or no in- Pain, depression, quality of Short term

injury months, strengthening exercises tervention. life, adverse effects (within 24 hours
Cochrane and 1 to 24 aimed at mobilising painful of last interven-
Injuries yr when re- shoulder joint. tion, i.e. post-
Group ported intervention)
and intermedi-
ate term (1 to 6
wk post-inter-
vention) and
long term (> 6
wk post-inter-
vention)
32

Collaboration.
Table 3. Characteristics of included reviews (Continued)
Brown 2010 Aug 2009 Primary Ongo- 12-wk walk or jog training Asked not to exercise Pain: menstrual disorders Ongoing - over
dysmenor- ing/not ap- programme at an intensity of during the experimental questionnaire (MDQ) score 3 menstrual cy-
Library
Cochrane
Cochrane rhoea in the propriate 70% to 85% of the HR range. period. cles
Menstrual majority (≥ Training for 3 days/wk and
Disorders 50%) of cy- duration of aerobic phase was
and Subfer- cles 30 minutes with 15-minute
tility Group warm-up and cool-down peri-
ods.
Better health.
Informed decisions.
Trusted evidence.
Busch 2007 Aug 2007 Fibromyal- Not report- Exercise-only interventions "Untreated." Pain, global wellbeing, ob- Post-interven-
gia ed included aerobic-only train- jectively measured physical tion (strength
Cochrane ing, strength-only training, function exercise 21 wk,
Muscu- flexibility-only training, or aerobic exer-
loskeletal mixed exercise-only interven- cise 6 to 23 wk)
Group tions.
Busch 2013 Mar 2013 Fibromyal- mean range Defined resistance training as Untreated control condi- Multi-dimensional func- Post-interven-
gia from 4 yrs exercise performed against tions (treatment as usu- tion, self-reported physical tion, follow-up
Cochrane (SD 3.1) to a progressive resistance on al, activity as usual, wait function, pain, tenderness, (12 wk) in 1
Muscu- 12 yrs (SD 4) a minimum of 2 days/wk (on list control, and place- muscle strength, adverse ef- study only
loskeletal non-consecutive days) with bo), other types of exer- fects, all-cause attrition
Group the intention of improving cise or physical activity
muscle strength, muscle en- interventions (e.g. aero-
durance, muscle power, or a bic, flexibility), and other
combination of these. resistance training inter-
ventions (head-to-head
comparisons).
Cramp 2013 Oct 2012 Rheumatoid Not report- Included pool-based therapy "Could have been place- Fatigue, pain, anxiety, de- Post-interven-
arthritis ed (twice/wk, moderate intensi- bo, an alternative inter- pression, disability, tender tion (only a sin-
Cochrane ty, music-paced), yoga (6 wk, vention (pharmacologi- and swollen joints, adverse gle time point
Muscu- twice/wk, 1.5-hour sessions), cal or non-pharmacolog- events analysed)
loskeletal dynamic strength training ical) or usual care."

Group (home-based after inpatient
programme, all main muscle
groups using dumbbells and
elastic bands), stationary cy-
cling (70% HRmax, 5 minute
excluding: 1-minute of rest,
increased duration), low-im-
pact aerobics (class at fitness
centre and video at home, in-
dividual HR targets), tai chi (1-
hour group sessions).
33

Collaboration.
Fransen May 2013 Hip OA Not report- Any land-based therapeutic Wait-list control, usual Self-reported pain, physi- post-interven-
2014 ed exercise regimens aiming to care, GP education. cal function, quality of life, tion (immediate
Library
Cochrane
relieve the symptoms of hip withdrawal or dropouts, ad- in 9/10 studies)
Cochrane OA, regardless of content, du- verse events follow-up 3 to 6
Muscu- ration, frequency, or intensi- months
loskeletal ty. This included any exercise
Group designed to improve mus-
cle strength, range of joint
movement or aerobic capac-
Better health.
Informed decisions.
Trusted evidence.
ity (or combinations of the
three). Programmes could be
designed and supervised by
physiotherapists or other pro-
fessionals, or provided as a
home programme with mini-
mal monitoring.
Fransen May 2013 Knee OA Often not "land-based therapeutic ex- No exercise: active (any Knee pain, self-reported Immediate-
2015 reported: ercise." Along with delivery no-exercise interven- physical function, quality of ly at the end
some less mode and content, treatment tion) or no treatment life of treatment
Cochrane than 1yr, 'dosage' (duration, frequency, (including waiting list). (post-treat-
Muscu- others over intensity) varied widely be- ment), 2 to
loskeletal 10yr tween studies. 6 months af-
Group ter cessation
of monitored
study treatment
and longer than
six months af-
ter cessation
of monitored
study treatment
Gross 2015a May 2014 Mechanical "Chron- Cervical stretch/ROM exercis- Wait list control. Pain intensity, function, Immediately

neck disor- ic" (not sub- es + cervical/scapulothoracic quality of life, global per- post-treatment
Cochrane ders acute or strengthening + static/dy- ceived effect, adverse ef- (≤ 1 day),
Back Group acute) namic cervical/shoulder sta- fects short-term fol-
bilisation. low-up (1 day to
3 months),
intermedi-
ate-term fol-
low-up (3
months up to,
but not includ-
ing, 1 yr), and
34

Collaboration.
long-term fol-
low-up (≥ 1 yr)
Library
Cochrane
Han 2004 Apr 2004 Rheumatoid Not report- Only trials of exercise pro- Not reported. Function, tender and Post-interven-
arthritis ed grammes with tai chi instruc- swollen joints, ROM, tion (8 to 10 wk)
Cochrane tion or incorporating princi- strength, enjoyment, with-
Muscu- ples of tai chi philosophy. drawals, adverse effects
loskeletal
Group
Better health.
Informed decisions.
Trusted evidence.
Hayden Sep 2004 Non-specif- Chronic, i.e. Exercise therapy defined as No exercise: no treat- Pain, functional ability, Earliest, 6 wk,
2005 ic low back longer than "a series of specific movement or placebo treat- work status, global assess- 6 months, 12
pain 12 wk: 5.6 yr ments with the aim of training ment, other conserva- ment, adverse events months
Cochrane (95% CI 3.4 or developing the body by a tive therapy, or another
Back Group to 7.8) routine practice or as physi- exercise group.
cal training to promote good
physical health;" only 54%
adequately described the ex-
ercise intervention.
Hurkmans Jun 2009 Rheumatoid 5 to 14 yr Dynamic exercise pro- Not reported Functional ability, aerobic Follow-up
2009 arthritis grammes - aerobic capacity capacity, muscle strength, (12 wk and 24
and muscle strength training; safety (pain and radiological months)
Cochrane short-term muscle strength damage)
Muscu- training (high quality); short-
loskeletal term dynamic exercise to im-
Group prove aerobic capacity (not
high methodological qual-
ity); exercise frequency of
at least 20 minutes twice a
week. Duration of exercise
programme at least 6 wk (du-
ration < 3 months was consid-

ered short-term; duration > 3
months was considered long-
term). Exercise programme
performed under supervision.
Aerobic exercise intensity at

least 55% of the maximum
HR; or intensity starting at
40% to 50% of the maximum
oxygen uptake reserve or HR
maximum reserve. Further-
more, the intensity was in-
creased up to 85% during the
35

Collaboration.
intervention. Progressively
strengthening exercise loads
Library
Cochrane
starting at 30% to 50% and in-
creasing to 80% of maximum
(defined as the percentage of
either 1 repetition maximum,
1 MVC, maximum speed, or as
maximal subjective exertion).
Better health.
Informed decisions.
Trusted evidence.
Koopman Jul 2014 Postpolio Not report- Exercise therapy (e.g. aerobic Placebo, usual care or Self-perceived activity lim- 3 and 6 months
2015 syndrome ed exercise, muscle strengthen- no treatment. itations, muscle strength,
(PPS) ing exercise, respiratory mus- muscle endurance, fatigue,
Cochrane cle training, warm climate pain, adverse events (minor
Neuromus- training, hydro training). and serious)
cular Group
Lane 2014 Sep-2013 intermittent not reported Any exercise programme Exercise was compared maximal walking time, pain- Post-interven-
claudication used in the treatment of in- to six different modes free walking time, pain-free tion, 3-month
Cochrane termittent claudication was of treatment, the most walking distance, maximum follow up, six-
Peripher- included, such as walking, common being usual walking distance, ankle month follow
al Vascu- skipping and running. Inclu- care or placebo. Two brachial index (ABI), peak up
lar Diseases sion of trials was not affect- early trials compared exercise calf blood flow,
Group ed by the duration, frequen- exercise with placebo mortality, amputation
cy or intensity of the exercise tablets but in more re-
programme but these issues cent studies usual care
were taken into account in was used as the control
the meta-analysis comparator. Exercise
was compared with the
following drug thera-
pies: antiplatelet agents
pentoxifylline, iloprost,
and vitamin E. One study
compared exercise with
pneumatic foot and calf

compression.
Lauret 2014 Jul 2013 Intermittent Not report- Supervised walking pro- Alternative exercise. Maximum walking distance n/a
claudication ed gramme needed to be super- (METs), pain-free walking
Cochrane vised at least twice a week for distance (METs), health-
Peripher- a consecutive 6 wk of train- related quality of life and
al Vascu- ing. functional impairment
lar Diseases
Group
36

Collaboration.
Regnaux Jun 2014 Hip or knee > 6 months High-intensity physical activi- Low-intensity physical Pain, physical function, Post-interven-
2015 OA ty or exercise programme. activity or exercise pro- quality of life, adverse ef- tion, interme-
Library
Cochrane
gramme fects (related to interven- diate term (6
Cochrane tion), severe adverse events to 12 months),
Muscu- and or withdrawal (due to inter- long-term (over
loskeletal vention) 12 months) fol-
Group control (no-exercise)
low-up
group in 1 study.
Better health.
Informed decisions.
Trusted evidence.
Saragiotto Apr 2015 Low back > 12 wk MCE: activation of the deep Placebo, no treatment, Pain intensity and disabili- Post-interven-
2016 pain trunk muscles, targeting the another active treat- ty, function, quality of life, tion, short term
restoration of control and co- ment, or when MCE was global impression of recov- (4 to 10 wk),
Cochrane ordination of these muscles. added as a supplement ery, return to work, adverse intermediate
Back and to other interventions. events and recurrence term (3 to 6
Neck Group When MCE was used in months), long
addition to other treat- term (12 to 36
ments, it had to rep- months)
resent at least 50% of
the total treatment pro-
gramme to be included.
Silva 2010 Jun 2009 Rheumatoid No studies Balance training (propriocep- No intervention or other ACR-50, pain, disease activ- n/a
arthritis found tive training). intervention. ity score (DAS), Health As-
Cochrane sessment Questionnaire
Muscu- (HAQ for function), gait, ad-
loskeletal verse effects, discontinua-
Group tion rate
van der Heij- May 2014 Adoles- 3 wk to 8 Exercise therapy for No treatment, placebo, Pain during activity, usual 4- to 12-wk fol-
den 2015 cents and months (as patellofemoral pain syn- or waiting list controls. pain, functional ability, re- low-up (short
adults with minimum drome; exercises could be This also included 'exer- covery term) and 16 wk
Cochrane patellofemoral require- performed at home or under cise therapy + another to 12 months
Bone, Joint pain ment); re- supervision of a therapist - intervention (e.g. taping) (long term)

and Mus- ported pain various descriptions in the in- versus the other inter-
cle Trauma 4 wk to 9 yr cluded trials, including knee vention alone (e.g. tap-
Group exercises, hip and knee exer- ing).'
cises, home exercises, super-
vised exercises, closed kinetic
chain, open kinetic chain.
Yamato Mar 2014 Low back Acute, suba- Explicitly stated as based on No intervention, place- Pain intensity, disability, Short term (4 to
2015 pain cute, chron- Pilates principles, or the ther- bo, or other interven- global impression of recov- 8 wk), interme-
ic (i.e. no apists who provided the inter- tions. ery, quality of life, return to diate term (3 to
Cochrane minimum) ventions had previous train- work, adverse effects 6 months)
Back Group ing in Pilates exercises or the
37

Informed decisions.

ACR: American College of Rheumatology; GP: general practitioner; HR: heart rate; MCE: motor control exercise; MET: metabolic equivalents; n/a: not applicable; OA: osteoarthritis;
therapists were described as
certified Pilates instructors
ROM: range of motion; wk: week; yr: year.

Collaboration.
Informed decisions.


Table 4. Further characteristics of included reviews
Review Number of trials included Total number of participants Gender distribution Participants ages
Bartels 6 (4 exercise vs no exercise) 800 (674 exercise vs no exercise) 50% to 86% Female Means ranged from
2007 66 to 71 yr
Bidonde 16 (9 exercise vs no exercise) 881 (519 exercise vs no exercise) 513 female, 6 male Means ranged from
2014 46.3 to 48.3 yr
Boldt 16 (3 exercise vs no exercise) 616 (149 exercise vs no exercise) 115 male, 41 female Range 19 to 65 yr and
2014 across 3 studies mean 35 to 45 yr
Brown 1 36 100% female Not reported

2010
Busch 34 (in meta-analysis - strength 2276 total 96.4% female when re- Range reported as
2007 training vs control: 2; ported (in 2197 partic- 27.5 to 60.2 yr
aerobic training vs control: 4) (in meta-analysis - strength: 47, ipants)
aerobic: 269)
Busch 5 studies as 7 publications (ex- 219 with fibromyalgia (exercise vs 100% female Not reported
2013 ercise vs control: 3 publications, control: 81)
2 studies)
Cramp 24 (only 6 using physical activity 2882 (physical activity interven- "A higher percentage "Mainly within the
2013 interventions) tions: 371) of females"… when fifth decade"
reported
Fransen 10 > 549 75% to 80% female 58 to 70 yr (means)

2014 when reported when reported
Fransen 54 5362 When reported 55% to When reported mean

2015 100% female age 60 to 70 yr
Gross 27 (16 chronic pain) 2485 Not reported Not reported

2015a
Han 2004 4 (3 RCTs). Pain not reported in 206 total; pain not reported in Not reported Range 38 to 72 yr
any included study any included study
Hayden 61 (43 chronic low back pain) 6390 (3907 chronic low back Chronic: 46% male Chronic: 42 yr (95%
2005 pain) (95% CI 39 to 52) CI 40 to 44)
Hurkmans 8 RCTs (5 exercise vs no-exer- 575 "Mainly female" 52 yr

2009 cise)
Koopman 13 (2 exercise vs no exercise) 675 (68 exercise vs no exercise) ˜ 25% male Mean 58 and 65 yr
2015 - 1 study used 3 arms (no treat-
ment in cold, exercise in cold, ex-
ercise in warm; we have excluded
the warm exercise arm as cannot
compare directly to the control)
Lane 2014 30 1822 total Not reported Mean > 65 yr
Lauret 5 (0 for exercise vs no exercise) 184 (0 for exercise vs no exercise) n/a n/a
2014
Collaboration.
Informed decisions.

Table 4. Further characteristics of included reviews (Continued)

Regnaux 6 (1 for exercise vs no exercise) 656 (102 for exercise vs no exer- 79 female 62.6 yr
2015 only 1 study that had a no exercise)
cise control
Saragiotto 29 (7 for exercise vs no exer- 2431 (671 for exercise vs no exer- "Mixed" Median 40.9 yr (IQR
2016 cise/minimal intervention) cise) 11.2) (range 20.8 to
54.8)
Silva 2010 None None n/a n/a
van der 31 (10 for exercise vs control) 1690 0% to 100% female; Mean 25 to 50 yr
Heijden equally distributed
2015 across range
Yamato 10 (6 exercise vs minimal inter- 478 (265 exercise vs control) 2 trials were all fe- Mean 38 yr (range 22
2015 vention (control)) male, the others in- to 50)
cluded both genders
CI: confidence interval; GP: general practitioner; IQR: interquartile range; OA: osteoarthritis; RCT: randomised controlled trial; ROM: range
of motion; wk: week; yr: year.

Table 5. Dose and duration of exercise interventions in included reviews
Review Duration Frequen- Intensity Duration Other description
cy (per ses-
(session)
sions per
day/wk/
month)
Bartels Not re- Not re- "Muscle mainte- Not re- No minimum requirement for inclusion.
2007 ported ported nance" and "range ported Actual intervention only reported by 2 of 6 included
of motion" studies.
Bidonde 17 wk 1 to 4/wk Very light (< 57% 45 min- No minimum requirement for inclusion.
2014 (range 4 HRmax) to vigor- utes
to 32) ous (95% HRmax), (range 30 None of the studies met the ACSM exercise guidelines
self-selected, and to 70) specified for aerobic or strength training. Only 1 study
not specified met the ACSM guidelines for flexibility training.
Boldt 12 wk to 9 2/day to Not reported Reported No minimum requirement for inclusion.

2014 months 2/wk for 1 study
only (90 to Stretching and strengthening exercises aimed at mobil-
120 min- ising painful shoulder joint.
utes)
Brown ≥ 12 wk 3/wk 70% to 85% HRR 1 hour No minimum requirement for inclusion.
2010
Busch 3 wk to 6 1 to 5/wk Not reported Not re- No minimum requirement for inclusion.
2007 months ported Assessed as whether they "met ACSM recommenda-
tions."
Busch 8 to 21 wk ≥ 2/wk > 4/10 RPE rat- 40 to 90 Assessed as whether they "met ACSM recommenda-
2013 (median ing progressing to minutes tions."
16 wk) 70% to 80% 1RM
Collaboration.
Informed decisions.

Table 5. Dose and duration of exercise interventions in included reviews (Continued)

Cramp 6 wk 2/wk "Low impact", 1 to 1.5 No minimum requirement for inclusion.
2013 (when re- "moderate", and hours,
ported) 70% HRmax when re-
ported
Fransen 6 to 12 wk 1 to 3/wk "Low intensity" to 30 to 60 No minimum requirement for inclusion.

2014 (median "max effort" minutes Intensity only reported in 2 of 10 studies.
8)
Fransen single ses- 1 to 5/wk "Moderate to 15 to 60 No minimum requirement for inclusion.

2015 sion to 30 moderately high minutes Varied in dose and duration.
months intensity"
Gross 2 wk to 3 5/wk to Low intensity 2 to 20 -

2015a months every minutes
15 min-
utes/day
Han 2004 8 to 10 wk 1 to 7/wk Tai chi = low inten- 1 to 1.5 No minimum requirement for inclusion.
(when re- (median sity hours
ported) 1/wk)
Hayden Not re- Not re- Not reported Not re- No minimum requirement for inclusion.
2005 ported ported ported Could not extract actual data.
Hurkmans ≥ 6 wk 2/wk Aerobic: ≥ 55% 20 min- -

2009 HRmax increasing utes
to 85% HRmax
strength: start
30% 1RM increas-
ing to 80% 1RM
Koopman 4 to 12 wk Daily to 3/ Reported in 1 45 min- No minimum requirement for inclusion.

2015 wk study: 50% to 70% utes
MVC 1 study: supervised progressive resistance training con-
sisting of 3 sets of 8 isometric contractions of the thumb
muscles.
1 study: combination of individual and group thera-

py with daily treatment in a swimming pool (45 min-
utes), physiotherapy, individually adapted training pro-
gramme.
Lane 2014 3 to 12 ≥ 2/wk "Variable" ˜ 60 min- No minimum requirement for inclusion.

months utes
Lauret ≥ 6 wk ≥ 2/wk Not reported Not re- No minimum requirement for inclusion.
2014 ported Must be supervised.
Regnaux 8 wk 3/wk Compared high vs 30 to 50 Every 2 wk 1RM was retested and increased by 5% as tol-
2015 low intensity vs minutes erated in each group.
control
Supervision: an experienced therapist.
3 arms (n=34 per arm): high intensity, low intensity, con-

trol (no exercise).
Saragiotto 20 days 1 to 5/wk Not reported 20 to 90 MCE is usually delivered in 1:1 supervised treatment ses-
2016 to 12 wk (median minutes sions, and sometimes involves ultrasound imaging, the
(median 12 ses- (median
Collaboration.
Informed decisions.

Table 5. Dose and duration of exercise interventions in included reviews (Continued)

8 wk (IQR sions (IQR 45 (IQR use of pressure biofeedback units or palpation to pro-
2.0)) 6.0)) 30) min- vide feedback on the activation of trunk muscles.
utes)
Silva 2010 ≥ 6 wk 2/wk Balance training ≥ 30 min- No studies found.

only utes
van der 3 to 16 wk 2/wk to Not reported Not re- No minimum requirement for inclusion.
Heijden daily ported Assessed by duration (< or > 3 months), frequency (sev-
2015 eral times, or once a week), medium (land or water), etc.
Yamato 10 to 2/wk Not reported 1 hour No minimum requirement for inclusion.

2015 90 days (mean Must be supervised (for the Pilates technique).
(mostly 8 session
wk) num-
ber 15.3,
range 6 to
30)
1RM: one repetition maximum; ACSM: American College of Sport Medicine; HRmax: maximum heart rate; HRR: heart rate reserve, IQR:
interquartile range; MCE: motor control exercise; MVC: maximum voluntary contraction; RPE: rating of perceived exertion; wk: week.

Collaboration.
Collaboration.
Table 6. Methodological quality of included reviews using the AMSTAR tool (Continued)
Review Criteria Total Total Total
"Y" "N" "n/
Library
Cochrane
1 2 3 4 5 6 7 8 9 10 11 a"
Bartels 2007 Y Y Y Y Y N Y Y Y N N 8 3 -
Bidonde 2014 N Y Y Y Y Y Y Y Y N N 8 3 -
Better health.
Informed decisions.
Trusted evidence.
Boldt 2014 Y Y Y Y Y Y Y Y Y Y N 10 1 -
Brown 2010 Y Y Y N Y Y Y Y n/a N N 7 3 1
Busch 2007 Y Y Y N Y Y Y Y Y N N 8 3 -
Busch 2013 Y Y Y Y Y Y Y Y Y Y N 10 1 -
Cramp 2013 Y Y Y Y Y N Y Y Y Y N 9 2 -
Fransen 2014 Y Y Y Y Y N Y Y Y N N 8 3 -
Fransen 2015 Y Y Y Y Y N Y Y Y N N 8 3 -
Gross 2015a Y Y Y Y Y N Y Y Y Y N 9 2 -
Han 2004 Y Y Y Y Y N Y Y N N N 7 4 -
Hayden 2005 Y Y Y Y Y N Y Y Y Y Y 10 2 -
Hurkmans 2009 Y Y Y Y Y Y Y Y Y N N 9 2 -
Koopman 2015 Y Y Y Y Y Y Y Y Y N Y 10 1 -

Lane 2014 Y Y Y Y Y Y Y Y N Y N 9 2 -
Lauret 2014 Y Y Y Y Y N Y Y Y Y N 9 2 -
Regnaux 2015 Y Y Y Y Y Y Y Y Y Y N 10 1 -
Saragiotto 2016 Y Y Y Y Y N Y Y Y Y N 9 2 -
Silva 2010 Y Y Y Y Y n/a n/a n/a n/a n/a Y 6 0 5

43

Collaboration.
Table 6. Methodological quality of included reviews using the AMSTAR tool (Continued)
van der Heijden 2015 Y Y Y Y Y Y Y Y Y N N 9 2 -
Library
Cochrane
Yamato 2015 Y Y Y Y Y N Y Y Y Y N 9 2 -
Total "Y" 20 21 21 19 21 10 20 20 17 10 3 - - -
Total "N" 1 - - 2 - 10 - - 2 10 18 - - -
Better health.
Informed decisions.
Trusted evidence.
Total "n/a" - - - - - 1 1 1 2 1 - - - -
N: no; n/a: not applicable; Y: yes; out of maximum summative score of 11.
Following arbitration, the authors removed the response "cannot answer" due to no responses as such.

Table 7. Risk of bias - studies assessed as low risk of bias (Continued)
Review Number Selection bias Perfor- Detec- Attrition Reporting Other bias
of stud- mance tion bias bias bias
ies in bias
assess-
ment Ran- Allo- Blinding Blinding Incom- Selective Sample size Other
dom se- cation of partici- of out- plete reporting biases
quence conceal- pants and come out- (studies) (stud-
gener- ment personnel assess- come ies)
ation (stud- (studies) ment data
(studies) (stud- (stud-
ies) ies) ies)
Bartels 2007 6 Not re- 3 Not re- 2 3 Not re- 2, n > 100 per arm -
ported ported ported
Bidonde 2014 9 5 3 2 8 8 5 1, n > 50 per arm 7

Boldt 2014 3 1 1 0 1 2 3 0 1
Brown 2010 1 0 0 0 0 1 1 1, n > 50 per arm -
Busch 2007 34 17 10 8 20 Unclear 32 5, n > 50 per arm -
Busch 2013 5 4 2 1 2 5 3 0, n > 50 per arm -

44

Collaboration.
Cramp 2013 7 5 2 0 Not re- 6 4 1
ported
Library
Cochrane
Fransen 2014 10 8 7 0 0 7 4 1, n > 50 per arm 7
Fransen 2015 54 40 22 3 4 29 10 5, total n > 200
Gross 2015a 16 8 8 1 0 11 0 0 11
Better health.
Informed decisions.
Trusted evidence.
Han 2004 4 2 0 0 0 0 Not re- 0
ported
Hayden 2005 43 27 22 Not re- 12 29 Not re- 10, total n > 100 -
ported ported +
5, total n > 200
Hurkmans 2009 8 8 1 - 4 5 - 1, total n > 200 1
Koopman 2015 2 1 0 0 0 0 0 0 1
Lane 2014 30 16 14 30 7 19 29 3, total n > 100
Lauret 2014 5 4 2 5 3 4 5 1, total n > 100 4
Regnaux 2015 1 1 0 0 1 0 0 1, total n > 100 1
Saragiotto 2016 7 5 4 1 1 2 7 1, total n > 100 7
1, total n > 200

Silva 2010 0 n/a n/a n/a n/a n/a n/a n/a n/a
van der Heijden 2015 10 8 6 0 0 6 9 2, total n > 100 10
Yamato 2015 9 5 5 2 7 7 9 0 9
Studies with low risk of bias (num- 264 165 112 53 72 144 121 total n > 100: 26 71
ber) total n > 200: 15
total n > 400: 0
45

Informed decisions.

27%
total n > 100: 10%
total n > 200: 6%
total n > 400: 0%
46%
55%
27%
20%
42%
63%
n: number of participants, n/a: not applicable.

-
Studies with low risk of bias (per-
centage)

Collaboration.
Informed decisions.


Table 8. Interpretation of results by original review authors
Review Review authors' conclusions Overview authors' assessment
of conclusions
Bartels 2007 "Aquatic exercise has some short-term beneficial effects on the condition Appropriate conclusions based
of OA patients with hip or knee OA or both. The controlled and randomised on available data. No mention
studies in this area are still too few to give further recommendations on of quality/risk of bias in conclu-
how to use this therapy... No long-term effects have been found." sions, though found to be high
quality in results section.
Bidonde 2014 "Low to moderate quality evidence relative to control suggests that aquat- Appropriate conclusions based
ic training is beneficial for improving wellness, symptoms, and fitness in on available data.
adults with fibromyalgia. Very low to low quality evidence suggests that
there are benefits of aquatic and land-based exercise, except in muscle
strength (very low quality evidence favoring land). No serious adverse ef-
fects were reported."
Boldt 2014 "Evidence is insufficient to suggest that non-pharmacological treatments Appropriate conclusions based
are effective in reducing chronic pain in people living with SCI. The bene- on available data.
fits and harms of commonly used non-pharmacological pain treatments
should be investigated in randomised controlled trials with adequate sam-
ple size and study methodology"
Brown 2010 "There is a lack of available evidence to support the use of exercise in the Review authors should not have
alleviation of symptoms associated with dysmenorrhoea. The limited evi- commented on lack of adverse
dence implies that there are no adverse effects associated with exercise." events as this was not reported
in the included study. The com-
ment on lack of adverse events
contravened present Cochrane
guidance.
Busch 2007 "There is moderate quality evidence that short-term aerobic training (at Appropriate conclusions based
the intensity recommended for increases in cardiorespiratory fitness) pro- on available data.
duces important benefits in people with FM in global outcome measures,
physical function, and possibly pain and tender points. There is limited ev-
idence that strength training improves a number of outcomes including
pain, global wellbeing, physical function, tender points and depression.
There is insufficient evidence regarding the effects of flexibility exercise.
Adherence to many of the aerobic exercise interventions described in the
included studies was poor."
Busch 2013 "We have found evidence in outcomes representing wellness, symptoms, Appropriate conclusions based
and physical fitness favoring resistance training over usual treatment and on available data.
over flexibility exercise, and favoring aerobic training over resistance train-
ing. Despite large effect sizes for many outcomes, the evidence has been
decreased to low quality based on small sample sizes, small number of
randomized clinical trials (RCTs), and the problems with description of
study methods in some of the included studies."
Cramp 2013 "There is some evidence that physical activity interventions ... may help to Appropriate conclusions based
reduce fatigue in RA. However, the optimal parameters and components on available data. However, no
of these interventions are not yet established." mention of quality/risk of bias
of studies in conclusion despite
low/unclear quality score in re-
sults and discussion sections.
No conclusions about effect on

pain (insufficient data).
Collaboration.
Informed decisions.

Table 8. Interpretation of results by original review authors (Continued)

Fransen 2014 "There is currently high-level evidence that land-based exercise will re- Evidence was good quality
duce hip pain, and improve physical function, among people with sympto- though sample sizes were of-
matic hip osteoarthritis." ten small (i.e. it is debatable if
this was high level evidence as
claimed by authors). Agree that
results demonstrate small but
significant benefit from interven-
tion.
Fransen 2015 "High-quality evidence suggests that land-based therapeutic exercise pro- Appropriate conclusions based
vides benefit in terms of reduced knee pain and quality of life and mod- on available data. May have been
erate-quality evidence of improved physical function among people with generous with quality assess-
knee OA… Despite the lack of blinding we did not downgrade the quality ment but this was stated in con-
of evidence for risk of performance or detection bias." clusions for transparency.
Gross 2015a "…there is still no high quality evidence and uncertainty about the effec- Appropriate conclusions based
tiveness of exercise for neck pain… Moderate quality evidence supports on available data.
the use specific strengthening exercises as a part of routine practice …
Moderate quality evidence supports the use of strengthening exercises,
combined with endurance or stretching exercises may also yield similar
beneficial results. However, low quality evidence notes when only stretch-
ing or only endurance type exercises … there may be minimal beneficial
effects for both neck pain and function."
Han 2004 "Tai chi appears to have no detrimental effects on the disease activity of Appropriate conclusions based
RA in terms of swollen/tender joints and activities of daily living…tai chi on available data. However, no
appears to be safe, since only 1 participant out of 121 withdrew due to ad- mention of quality/risk of bias in
verse effects and withdrawals were greater in the control groups than the conclusion despite very low qual-
tai chi groups." ity score in results section.
Hayden 2005 "Evidence from randomized controlled trials demonstrates that exercise Appropriate conclusions based
therapy is effective at reducing pain and functional limitations in the treat- on available data. However, no
ment of chronic low-back pain, though cautious interpretation is required mention of quality/risk of bias of
due to limitations in this literature." studies in conclusion despite low
quality score in results and dis-
cussion sections.
Hurkmans 2009 "Short-term, land-based dynamic exercise programs have a positive ef- Appropriate conclusions based
fect on aerobic capacity (aerobic capacity training whether or not com- on available data. However, no
bined with muscle strength training) and muscle strength (aerobic capaci- mention of quality/risk of bias of
ty training combined with muscle strength training) immediately after the studies in conclusion.
intervention, but not after a follow-up period. Short-term, water-based dy-
namic exercise programs have a positive effect on functional ability and No conclusions regarding pain
aerobic capacity directly after the intervention but it is unknown whether severity.
these effects are maintained after follow-up. Long-term, land-based dy-
namic exercise programs (aerobic capacity and muscle strength training)
have a positive effect on functional ability, aerobic capacity, and muscle
strength immediately after the intervention but it is unknown whether
these effects are maintained after follow-up... Based on the evidence, aer-
obic capacity training combined with muscle strength training is recom-
mended for routine practice in patients with RA."
Koopman 2015 "Data from two single trials suggested that muscle strengthening of Appropriate conclusions based
thumb muscles (very low-quality evidence) ... are safe and beneficial for on available data.
improving muscle strength ... with unknown effects on activity limita-
tions."
"We found evidence varying from very low quality to high quality that ...
rehabilitation in a warm or cold climate are not beneficial in PPS."
Collaboration.
Informed decisions.

Table 8. Interpretation of results by original review authors (Continued)

"Due to a lack of good-quality data and randomised studies, it was impos-
sible to draw definitive conclusions about the effectiveness of interven-
tions in people with PPS."
Lane 2014 "… Exercise therapy should play an important part in the care of selected Appropriate conclusions based
patients with intermittent claudication, to improve walking times and dis- on available data. However, no
tances. Effects were demonstrated following three months of supervised mention of quality/risk of bias of
exercise although some programmes lasted over one year." studies in conclusion.
No conclusions regarding pain

severity.
Lauret 2014 "There was no clear evidence of differences between supervised walking Appropriate conclusions based
exercise and alternative exercise modes in improving the maximum and on available data. However, no
pain-free walking distance of patients with intermittent claudication…. mention of quality/risk of bias of
The results indicate that alternative exercise modes may be useful when studies in conclusion (in discus-
supervised walking exercise is not an option for the patient." sion).
Regnaux 2015 "We found very low- to low-quality evidence for no important clinical ben- Appropriate conclusions based
efit of high-intensity compared to low-intensity exercise programs in im- on available data. This overview
proving pain and physical function in the short term.... The included stud- has only used one study of the six
ies did not provide any justification for the levels of intensity of exercise included as it alone included a
programs. No authors reported evidence for the minimal and maximal in- control group, for which we could
tensity that could be delivered." not extract data as the control
comparison was not used in the
analysis by the review authors.
Saragiotto 2016 "There is very low to moderate quality evidence that MCE has a clinical- Appropriate conclusions based
ly important effect compared with a minimal intervention for chronic low on available data.
back pain... As MCE appears to be a safe form of exercise and none of the
other types of exercise stands out, the choice of exercise for chronic low
back pain should depend on patient or therapist preferences, therapist
training, costs and safety."
Silva 2010 "We were not able to provide any evidence to support the application of Appropriate conclusions based
balance exercises (proprioceptive training) alone in patients with RA." on available data (no included
studies).
van der Heijden "This review has found very low quality but consistent evidence that exer- No subgroup analysis to differen-
2015 cise therapy for patellofemoral pain syndrome (PFPS) may result in clini- tiate between acute, subacute,
cally important reduction in pain and improvement in functional ability." and chronic pain made it difficult
to extract appropriate data for
this review.
Yamato 2015 "No definite conclusions or recommendations can be made as we did not Appropriate conclusions based
find any high quality evidence for any of the treatment comparisons, out- on available data.
comes or follow-up periods investigated. However, there is low to moder-
ate quality evidence that Pilates is more effective than minimal interven- There was no subgroup analysis
tion in the short and intermediate term as the benefits were consistent for to differentiate between acute,
pain intensity and disability, with most of the effect sizes being considered subacute, and chronic pain made
medium." it difficult to extract appropriate
data for this review (one included
study had subacute back pain (>
6 weeks), all others were chron-
ic back pain (> 12 weeks)) but re-
sults are presented altogether as
chronic pain.
FM: fibromyalgia; MCE: motor control exercise; OA: osteoarthritis; PPS: postpolio syndrome; RA: rheumatoid arthritis; SCI: spinal cord injury.
Collaboration.
Informed decisions.


Table 9. Pain severity
Review Number Baseline pain Post-intervention reported Follow-up Overall comment/state-
of trials score result or change data (or if ment
(and par- only one data point report-
ticipants) ed in review)
assessing
'pain sever-
ity'
Bartels Hip + knee Control base- Hip + knee OA Hip + knee OA Statistically significant post-
2007 OA: line: intervention in hip + knee OA
A minor effect of a 3% ab- Follow-up at 6 group, but not clinically sig-
(os- Post-inter- Hip + knee OA solute reduction (0.6 few- months: SMD 0.11 nificant.
teoarthri- vention: 4 er points on WOMAC 0 to 20 (95% CI -0.12 to
tis) (638) WOMAC 0 to 20 scale) and 6.6% relative re- 0.33) (ns) Knee-only OA had moderate
(2 studies): 9.10 duction to large effect size (statisti-
Follow-up: 1 (SD 3.14) No difference cally significant) immediately
(310) SMD 0.19 (95% CI 0.04 to post-intervention.
VAS 0 to 100 (1 0.35) (P = 0.02) Hip only
Hip only: study): 55.3 (SD
24.6) Knee only SMD 1.00 (95% CI
follow-up: 1 -0.04 to 2.04) (P =
(17) HAQ 0 to 3 (1 SMD 0.86 (95% CI 0.25 to 0.06, ns)
study): 1.05 (SD 1.47)
Knee only: 0.61)
(P = 0.005)
post-inter- Hip only
vention: 1 Absolute difference 12% (1.2
(46) VAS 0 to 100 (1 fewer points on a 0 to 10
study): 56 (SD scale)
21.89)
Relative change 22% im-
Knee only provement
VAS 0 to 10 (1
study): 5.6 (SD
1.4)
Bidonde Post-inter- Weighted mean On 100-point scale: 3 studies at 12, 48, "We found a moderate effect
2014 vention: 7 score at base- or 52 weeks' post- favouring the aquatic exer-
(382) line (all partici- MD -6.59 (95% CI -10.71 to intervention cise training for pain" …"sim-
(fi- pants): 69.59 -2.48) could not be com- ilar improvements in pain
bromyal- median value bined. in the low pain groups (SMD
gia) SMD -0.53 (95% CI -0.76 to
for pain was 2 studies showed -0.60, 95% CI -0.98 to -0.23)
-0.31)
70.9 in stud- SMD favouring in- and in the high pain groups
Absolute difference -7% (95%
ies comparing tervention at fol- (SMD -0.57, 95% CI -1.11 to
CI -11 to -3)
aquatic training low-up. -0.03)."
to control NNTB 5 (95% CI 3 to 8)
Among the major wellness
outcomes, none of the out-
comes met the threshold for
clinically relevant differences
(15%).
Boldt Post-inter- WUSPI score WUSPI change score: 1 study at 4 weeks: "All three studies were
2014 vention: 3 22.6 (exercise fraught with high overall
(149) group) to 11.05 Exercise group: -7.7 (SD VAS (0 to 10): risk of bias. In particular, the
(spinal (control group) 19.01) -2.50 (95% CI comparison with 'no treat-
cord in- in 1 group at -3.48 to -1.52) (P < ment' or waiting lists as con-
jury) Control group: 12.8 (SD 0.00001)
baseline trol interventions likely leads
12.74)
to an overestimation of the
Collaboration.
Informed decisions.

Table 9. Pain severity (Continued)

Not reported SF-36 (pain experience): -1.9 WUSPI: -26.40 effectiveness of the exercise
for 2 studies (95% CI -3.4 to -0.4) favoured (95% CI -37.62 to programmes provided in
exercise (P = 0.01) -15.18favoured these studies. Consequently,
exercise (P < no conclusion on their effec-
VAS (0 to 10): MD -2.8 (95% CI 0.00001) tiveness can be drawn."
-3.77 to -1.83) favoured exer-
cise (P < 0.00001)
Busch Strength Control base- Aerobic training: SMD 0.65 n/a ">30% improvement was
2007 training: 1 line: (95% CI -0.09 to 1.39) (ns) seen in the strength train-
(21) Aerobic ing group as compared to an
(fi- training: 3 Aerobic: 6.1/10 Weighted absolute change untreated control group in
bromyal- (183) (VAS) (SD 1.97) 13% (1.3 cm lower on 10-cm pain."
gia) scale)
Strength: Aerobic training led to an im-
35/100 (VAS) Relative change 21% provement of 1.3/10.
(SD 19)
Strength training: SMD 3.00
(95% CI 1.68 to 4.32) (ns)
Weighted absolute change

49% (49 points lower on 100-
point scale)
Relative change 140%, NNTB

2
Busch Post-inter- Not reported Change score on VAS (in cm): 8 weeks: MD -0.68 > 30% improvement post-in-
2013 vention: 2 - change data MD -3.30 (95% CI -6.35 to (95% CI -1.62 to tervention.
(81) only -0.26) (P = 0.03) 0.26) (ns)
(fi-
bromyal- Follow-up SMD -1.89 (95% CI -3.86 to 16 weeks: MD
gia) at 8 weeks, 0.07) -1.79 (95% CI
16 weeks, 28 -2.70 to -0.88) (P <
weeks: 1 (60) Relative % change 44.6% 0.001)
(95% CI 3.5 to 85.9) favoured
exercise 28 weeks: MD
-0.85 (95% CI -1.77
NNTB 2 (95% CI 1 to 34) to 0.07) (P = 0.07,
ns)
Overall (n = 180):
MD -1.12 (95% CI
-1.65 to -0.58) (P <
0.0001)
Cramp 4 (not re- Not reported In narrative only - Harkcom Not reported "Improvement over time"
2013 ported) 1985: statistics not report- with "significant improve-
ed separately for pain data, ment in 24 months."
(rheuma- but reported as improvement
toid over time; Hakkinen 2003: No actual data available.
arthritis) "stat significant improve-
ment in 24 months"; Evans
2012 and Wang 2008: no sta-
tistically significant effects
Fransen End of treat- Not reported; End of treatment: SMD -0.38 3 to 6 months: "Small to moderate" statis-
2014 ment: 9 (549) land based ex- (95% CI -0.55 to -0.20) "small SMD -0.38 (95% tically significant improve-
ercise vs no ex- to moderate" favoured exer- CI -0.58 to -0.18) ment, but only mild pain at
(OA) 3 to 6 ercise: mean cise (P < 0.0001) "small to moder- baseline.
months: 5 pain in control ate" favoured ex-
(391) group ˜ 29/100 ercise (P = 0.0002)
Collaboration.
Informed decisions.


(based on 9
studies' control
values)
Fransen End of treat- Not reported; Land-based exercise vs no ex- 2 to 6 months: Absolute improvement of
2015 ment: 44 land-based exercise: SMD -0.24 (95% 12/100 post-intervention
(3537) ercise vs no ex- CI -0.35 to -0.14) (statistically significant).
(OA) ercise: mean Mean pain in intervention favoured exercise
Follow-up (2 pain in control groups was 0.49 SDs lower (P < 0.00001)
to 6 months): group 44/100 (95% CI 0.39 to 0.59 lower).
12 (1468) (based on 1 This translates to an absolute > 6 months: SMD
study control mean reduction of 12 points -0.52 (95% CI -1.01
Follow-up (> (95% CI 10 to 15) compared to -0.03) favoured
values)
6 months): 8 with control group on a 0 to exercise (P = 0.04)
(1272) 100 scale.
SMD -0.49 (95% CI -0.39 to

-0.59) (P < 0.00001)
Absolute reduction 12% (95%

CI 10% to 15%)
Relative change 27% (95% CI

21% to 32%)
NNTB 4 (95% CI 3 to 5)
Gross 12-week Not reported, 12 weeks: pooled MD -14.90 24 weeks: pooled 2 trials showed a moderate
2015a treatment: 2 but control (95% CI -22.40 to -7.39) MD -10.94 (95% (statistically significant) re-
(147) scores at end of favoured exercise (P = 0.0001) CI -18.81 to -3.08) duction in pain post-inter-
(mechan- treatment 40 to favoured exercise vention (14.9/100).
ical neck 24 week (or 60/100 (moder- (P = 0.0064)
disorders) 12-week ate pain)
treatment +
12-week fol-
low-up): 2
(140)
Hayden Earliest fol- "Chronic Earliest: MD -10.20 (95% CI Short term: MD Reduction of ˜ 10/100 at ear-
2005 low-up: 8 group" at base- -19.09 to -1.31) (P = 0.02) -8.58 (95% CI liest measurement point.
(370) line: mean -18.46 to 1.29) (P =
(low back 46/100 (95% CI 0.09, ns)
pain) Follow-up 41 to 50) (mod-
(time since erate pain) Intermediate
randomisa- term: MD -12.48
tion) (95% CI -22.69 to
-2.27) (P = 0.02)
Short term
(6 weeks): 6 Long term: MD
(268) -3.93 (95% CI -9.89
to 2.02) (P = 0.2,
Intermedi- ns)
ate term (6
months): 5
(249)
Long term
(12 months):
2 (126)
Collaboration.
Informed decisions.


Hurkmans 4 studies (to- Not reported Short-term (12 weeks): Long-term (24 No significant difference be-
2009 tal 188 par- months) tween control and interven-
ticipants) Short-term land-based (aer- land-based (aero- tion.
(rheuma- in different obic and strength training) bic and strength
toid categories SMD -0.53 (95% CI -1.09 to training)
arthritis) (results not 0.04)
combined) SMD 0.35 (95% CI
Short-term land-based (aero- -0.46 to 1.16)
bic only)
SMD -0.27 (95% CI -0.79 to
0.26)
Short-term water-based SMD

0.06 (95% CI -0.43 to 0.54)
Koopman 1 (55) Not reported, 3 months post-intervention: n/a No significant effect/no dif-
2015 but control ference between groups.
scores at end VAS (0 to 100): MD 11.00 (95%
(postpo- of treatment CI -0.98 to 22.98) (P = 0.072)
lio syn- mean 44 (SD
drome) 24) on a 0 to
100 scale (mod-
erate pain)
Regnaux Only 1 study Not reported Post-intervention: WOMAC (0 n/a Actual individual study data
2015 that had a to 20) was extracted (where possi-
no-exercise ble) instead of pooled MD or
(OA) control: Change data presented for SMD due to comparison this
high- vs low-intensity groups overview wishes to make (ex-
1 (68) - ex- only, not compared to con- ercise vs no-exercise only).
cluded data trol
for control Could not extract exercise vs
(no exercise) control data.
from analy-
sis (n = 34)
Saragiotto Short term (< Not reported, Short term: MD -10.01 (95% Intermediate Medium effect size favouring
2016 3 months): 4 but control CI -15.67 to -4.35) favoured term: MD -12.61 exercise at all follow-up as-
(291) scores at fol- exercise (P < 0.001) (95% CI -20.53 to sessments (moderate quality
(low back low-up range -4.69) favoured ex- evidence at short- and long-
pain) Intermediate 25 to 56/100 ercise (P = 0.002) term, low quality evidence at
term (3 to 12 (mild-moderate intermediate term).
months): 4 pain) Long term: MD
(348) -12.97 (95% CI Clinically important effect.
-18.51 to -7.42)
Long term (> favoured exercise
12 months): (P < 0.001)
3 (279)
van der 3 studies Not reported, Short-term (4 to 8 weeks): "Long term" (16 Reduction in pain of 4/10 at
Heijden with pain > but control weeks) VAS (0 to 16 weeks' follow-up.
2015 3 months scores at fol- MD for usual pain in the exer- 10): MD -4.42 (95%
(135 partic- low-up range cise group was 0.93 (95% CI CI -7.75 to -0.89)
(patellofemoral
ipants), 2 2.1 to 6.0/10 1.60 to 0.25) SDs lower favoured exercise
pain syn- studies used (mild-moderate (P = 0.01)
drome) SMD -0.93 (95% CI -1.60 to
in analysis pain)
-0.25) (P = 0.008)
(41 partici-
pants)
Collaboration.
Informed decisions.


Long-term
follow-up: 1
(94)
Yamato Short term: 6 Not reported, Short-term follow-up (< 3 Intermediate term "Low quality evidence
2015 (265) but control months): (3 to 12 months): (downgraded due to impreci-
scores at earli- MD -10.54, (95% CI sion and risk of bias) that Pi-
(low back Intermediate est follow-up MD -14.05 (95% CI -18.91 to -18.54 to -2.62) (P lates reduces pain compared
pain) term: 2 (148) range 18 to -9.19) (P < 0.001) = 0.009) with minimal intervention at
52/100 (mild- short-term follow-up, with a
moderate pain) medium effect size...
intermediate-term follow-up,
two trials, provided moder-
ate quality evidence (down-
graded due to imprecision)
that Pilates reduces pain
compared with minimal in-
tervention, with a medium
effect size"
CI: confidence interval; HAQ: Health Assessment Questionnaire; MD: mean difference; n/a: not applicable; NNTB: number needed to
treat for an additional beneficial outcome; ns: not significant; OA: osteoarthritis; SD: standard deviation; SF-36: 36-item Short Form;
SMD: standardised mean difference; VAS: visual analogue score; WOMAC: Western Ontario and McMaster Universities Osteoarthritis Index;
WUSPI; Wheelchair User Shoulder Pain Index.

Collaboration.
Collaboration.
Table 10. Physical function
Review Outcome Number of trials (and Post-intervention result (or if Short-term fol- Intermedi- Long-term Overall com-
measure participants) used in only 1 result reported) low-up (or if only ate-term follow-up ment/statement
Library
Cochrane
analysis 1 follow-up point follow-up
reported)
Bartels 2007 Self-report- Post-intervention Function (hip + knee): SMD 0.26 Hip only Hip + knee n/a Function was signif-
ed function (95% CI 0.11 to 0.42) favoured ex- Disability, SMD 1.00 (6 months) icantly improved in
(OA) (WOMAC Hip + knee ercise (P < 0.001) (95% CI -0.04 to people with hip + knee
function: 4 (648) Function,
Better health.
Informed decisions.
Trusted evidence.
and HAQ) 2.04) favoured ex- OA immediately post-
and walking walking ability: 2 (355) Walking (hip + knee): SMD 0.18 ercise (P = 0.06, ns) SMD 0.10 intervention only -
ability, and (95% CI -0.03 to 0.39) favoured (95% CI small effect size only.
Hip only function: 1 exercise (P = 0.08, ns) -0.12 to
DRI
(28) 0.33) (ns)
Function (hip only): SMD 0.76
Follow-up function hip (95% CI -0.02 to 1.53) favours ex-
+ knee: 1 (306) ercise (P = 0.06, ns)
hip only: 1 (17)
Bidonde Self-report- 5 (285) MD -4.35 (95% CI -7.77 to -0.94) n/a n/a n/a Small difference (im-
2014 ed physical provement) in aquatic
function (0 SMD -0.44 (95% CI -0.76 to -0.11) exercise group.
(fibromyal- to 100 scale)
gia) Absolute difference -4 (95% CI -8 Among the major well-
to -1) ness outcomes, none
of the outcomes met
NNTB 6 (95% CI 3 to 22)
the threshold for clin-
ically relevant differ-
ences (15%).
Busch 2007 Physical Aerobic: 4 (253) Aerobic: SMD 0.66 (95% CI 0.41 n/a n/a n/a Function was signifi-
function to 0.92) favoured exercise (P < cantly improved from
(fibromyal- Strength: 2 (47) 0.0001) aerobic exercise train-
gia) ing, strength training
Strength: SMD 0.52 (95% CI -0.07

neared significance.
to 1.10) favoured exercise (P =
0.08, ns) Moderate effect size.
Busch 2013 HAQ and 3 (107) Change score MD -6.29 (95% CI n/a n/a n/a Significantly
SF-36 for -10.45 to -2.13) favoured exercise favourable effect of ex-
(fibromyal- function (P < 0.01) ercise.
gia)
Cramp 2013 Disability 4 (not reported) n/a n/a n/a n/a "Studies investigat-
ing hydrotherapy and
(rheumatoid tai chi demonstrated
arthritis) statistically significant
55

Collaboration.
Table 10. Physical function (Continued)
improvements in the
intervention arm com-
Library
Cochrane
pared to the control
arm between baseline
and follow-up.
The studies investigat-
ing strength training
and Ivengar yoga did
not demonstrate a sta-
Better health.
Informed decisions.
Trusted evidence.
tistically significant
difference between
study arms."
Fransen Physical Post-intervention: 9 SMD -0.30 (95% CI -0.54 to -0.05) SMD -0.37 (95% n/a n/a Statistically signifi-
2014 function (521) "significant benefit" favoured ex- CI -0.57 to -0.16) cant, but small effect
ercise (P = 0.02) favoured exercise size only.
(OA) Follow-up (3 to 6 The demonstrated effect size for (P < 0.001)
months): 5 (365) exercise was equivalent to an im-
provement of physical function
of 7 points (95% CI 1 to 12) on a
0 to 100 scale compared with a
control group
Fransen Physical Post-intervention: 44 SMD -0.52 (95% CI -0.64 to -0.39) SMD -0.15 (95% SMD -0.57 n/a Significant effect from
2015 function (3913) favoured exercise (P < 0.0001); an CI -0.26 to -0.04) (95% CI exercise at every fol-
improvement of 10 points (95% favoured exercise -1.05 to low-up point.
(OA) Follow-up (2 to 6 CI 8 to 13) on a 0- to 100-point (P = 0.008) -0.10)
months): 10 (1279) scale favoured ex- Moderate effect size
ercise (P = at short- and long-
Follow-up (> 6 term follow-up, but
0.02)
months): 8 (1266) only small effect at in-
termediate-term fol-
low-up.

Gross 2015a Physical 12 wk: 2 (147) 12 wk treatment: pooled SMD 24 wk treatment n/a n/a 2 trials showed a mod-
function -0.50 (95% CI -1.04 to 0.03) (or 12 wk' treat- erate (statistical) im-
(mechanical 24 wk: 2 (140) favoured exercise (P = 0.07, ns) ment + 12 wk fol- provement in function.
neck disor- low-up): pooled
ders) SMD -0.40 (95%
CI -0.74 to -0.06)
favoured exercise
(P = 0.02)
Han 2004 Functional Function: 2 (52) Function: MD 0.01 (95% CI -2.94 n/a n/a n/a No significant effect.
assessment to 2.97) (ns)
Walk test: 2 (48)
56

Collaboration.
(rheumatoid and 50-feet Walk test: MD 0.35 seconds (95%
arthritis) walk test CI -1.14 to 1.84) (ns)
Library
Cochrane
Hayden Function Earliest: 7 (337) Earliest: MD -2.98 (95% CI -6.48 to Short term: MD Intermedi- Long term: Favoured exercise
2005 0.53) favoured exercise (P = 0.09, -3.03 (95% CI -6.35 ate term: MD -4.22 from the earliest mea-
Short term: 6 (268) ns) to 0.53) favoured MD -3.84 (95% CI sure, but only reached
(low back exercise (P = 0.07, (95% CI -7.99 to statistical significance
pain) Intermediate term: 4
ns) -7.06 to -0.46) at intermediate and
(216)
Better health.
Informed decisions.
Trusted evidence.
-0.61) favoured ex- long term after ran-
Long term: 2 (126) favoured exercise (P = domisation.
ercise (P = 0.03)
0.02)
Hurkmans Functional Land-based aerobic: 2 n/a Short-term training n/a n/a No significant differ-
2009 ability (66) (12 wk) ence between con-
trol and intervention
(rheumatoid Land-based aerobic + Land-based aero- groups.
arthritis) strength: 2 (74) bic only training
SMD 0.03 (95% CI
-0.46 to 0.51) (ns)
Land-based aero-
bic and strength
training SMD -0.4
(95% CI -0.86 to
0.06) (ns)
Koopman Muscle Strength: 1 (10) Isometric muscle strength Activity limitation: n/a n/a Activity limitation:
2015 strength; (postintervention): MD 39.00% 6-months post-in- favoured intervention
and activ- Activity limitation: 1 (95% CI 6.12 to 71.88) tervention: at both assessment
(postpolio ity limita- (53) points.
syndrome) tion (Sun- Activity limitation: 3 months' ADL-index: MD
naas ADL-in- postintervention: -2.90 (95% CI -4.73 "The baseline imbal-

dex range 0 to -1.07) ance in favour of the
ADL-index: MD -2.70 (95% CI -4.53 usual care group prob-
to 36; River-
to -0.87) RMI: MD -1.80 (95% ably biased these re-
mead Mo-
CI -3.19 to -0.41) sults."
bility Index Rivermead Mobility Index (RMI):
(RMI) range MD -1.50 (95% CI -2.93 to -0.07)
0 to 15)
Lane 2014 Maximal Post-intervention Time: MD 4.51 minutes (95% CI Time: MD 6.05 min- Time: MD n/a Objectively measured
walking Walking time: 12 (577) 3.11 to 5.92) favoured exercise (P utes (95% CI 5.47 to 3.20 min- walking time and dis-
(intermit- time and < 0.00001) 6.62) favoured ex- utes (2.04 tance showed signifi-
tent claudi- maximal Walking distance: 9 ercise (P < 0.00001) to 4.36) cant improvement.
cation) walking dis- (480) favoured ex-
57
tance

Collaboration.
3-month follow-up Distance: 108.99 m (95% CI 38.20 Distance: MD ercise (P <
Walking time: 5 (174) to 179.78) favoured exercise (P = 104.46 m (95% CI 0.0001)
Library
Cochrane
0.003) -64.33 to 273.24)
Walking distance: 3 favoured exercise Distance:
(116) (ns) MD 138.36
m (95%
6-month follow-up CI 22.39
Walking time: 4 (295) to 254.34)
Walking distance: 3 favoured ex-
Better health.
Informed decisions.
Trusted evidence.
(156) ercise (P =
0.02)
Lauret 2014 Maximal No relevant studies n/a n/a n/a n/a No relevant studies.
walking
(intermit- time (mins)
tent claudi- and maxi-
cation) mal walk-
ing distance
(metres)
Regnaux WOMAC (0 1 (68) - excluded con- n/a n/a n/a n/a Could not extract ex-
2015 to 68) dis- trol (no-exercise data: ercise vs control data
ability scale, n = 34) - data presented for
(OA) and muscle high vs low intensity
strength groups only, not com-
pared to control.
Saragiotto Disability Short-term follow-up - MD -8.63 (95% CI MD -5.47 MD -5.96 Small effect sizes,
2016 (Oswestry (< 3 months): 5 (332) -14.78 to -2.47) (P < (95% CI (95% CI favoured exercise.
Disability In- 0.01) -9.17 to -9.81 to
(low back dex, Roland Intermediate term (3 -1.77) (P = -2.11) (P = Short term: CI included
pain) Morris Dis- to 12 months): 4 (348) 0.004) 0.002) a clinically important
ability Ques- effect.
Long term (> 12

tionnaire)
months): 3 (279)
Silva 2010 HAQ func- No studies found n/a n/a n/a n/a No studies found.
tion
(rheumatoid
arthritis)
van der Heij- Functional Short-term follow-up: n/a Short-term (4 to 8 n/a SMD 1.62 Significant effect of ex-
den 2015 ability 7 (483) wk): (95% CI ercise.
0.31 to 2.94)
Long-term follow-up: 3 SMD 1.10 (95% favoured ex-
(274) CI 0.58 to 1.63)
58

Collaboration.
(patellofemoral favoured exercise ercise (P = Very large effect size at
pain syn- (P < 0.0001) 0.02) short- and long-term
Library
Cochrane
drome) follow-up.
Yamato Disability Short-term (< 3 n/a MD -7.95 (95% CI MD -11.17 n/a "Low quality evidence
2015 (all mea- months) follow-up: 5 -13.23 to -2.67) (P = (95% CI (downgraded due to
sures con- (248) 0.003) -18.41 to imprecision and incon-
(low back verted to 0 -3.92) (P = sistency) that Pilates
pain) to 100 scale) -Intermediate-term 0.0025) improves disability at
Better health.
Informed decisions.
Trusted evidence.
(3 to 12 months) fol- short-term follow-up
low-up: 2 (146) compared with mini-
mal intervention, with
a small effect size ....
intermediate-term fol-
low-up, two trials pro-
vided moderate qual-
ity evidence (down-
graded due to impre-
cision) of a significant
effect in favour of Pi-
lates, with a medium
effect size"
ADL: activities of daily living; CI: confidence interval; DRI: Disability Rating Index; HAQ: Health Assessment Questionnaire; MD: mean difference; n/a: not applicable; NNTB: number
needed to treat for an additional beneficial outcome; ns: not significant; OA: osteoarthritis; SF-36: 36-item Short Form; SMD: standardised mean difference; wk: week; WOMAC:
Western Ontario and McMaster Universities Osteoarthritis Index,


59

Informed decisions.


Table 11. Psychological function
Review Outcome measure Number of tri- Outcome result (postintervention Fol- Additional state-
als (and partic- or if only one measurement point) low-up ment/comment
ipants) report-
ing psychological
function
Mental health
Bartels - 4 studies SMD 0.16 (95% CI 0.01 to 0.032) No signif- Very small effect
2007 favoured aquatic exercise icant dif- size postinterven-
ference at tion.
6 months,
1 study
Busch SF-36 - Mental health 1 study - n/a No group differ-

2013 scale ences.
Bidonde SF-36 - mental 4 studies, n = 243 MD -3.03 (95% CI -8.06 to 2.01) n/a No effect.
2014 Health scale
SF-12 - Mental Health

scale
Anxiety
Cramp Brief Symptom In- 1 study "No significant effect" n/a -

2013 ventory
Depression
Boldt CES-D 1 study, n = 34 MD -6.0 (95% CI -15.87 to 3.87) (P = n/a No effect.

2014 0.23)
Busch HADS - Depression 1 study, n = 21 MD -3.70 (95% CI -6.37 to -1.03) n/a Significant effect,
2013 favoured resistance
Beck Depression In- Relative difference 57% training.
dex
Cramp CES-D Not reported "Variable effect" reported in text on- n/a -
2013 ly
CES-D: Centre for Epidemiological Studies-Depression; CI: confidence interval; HADS: Hospital Anxiety and Depression Scale; MD: mean
difference; n: number of participants; n/a: not applicable; SF-12: 12-item Short Form; SF-36: 36-item Short Form; SMD: standardised mean
difference.

Table 12. Quality of life
Review Outcome mea- Number of trials Outcome result Additional statement/comment
sure (and participants)
reporting Quality of
Life (QoL)
(Health-related) Quality of Life
Collaboration.
Informed decisions.

Table 12. Quality of life (Continued)

Bartels QoL: SF-12 Hip + knee OA (post- Hip + knee (post-intervention): SMD Significantly favoured aquatic ex-
2007 (Physical), intervention): 3 stud- 0.32 (95% CI 0.03 to 0.61) (P = 0.028) ercise post-intervention in hip +
PQoL, EuroQoL ies, n = 599 knee OA.
Hip only (post-intervention): SMD 0.76
Hip only OA (post-in- (95% CI -0.02 to 1.53) (ns) Small effect size only (when sta-
tervention): 1 study, tistically significant).
n = 28 Hip only (follow-up): SMD 1.00 (95% CI
-0.04 to 2.04) (ns)
Hip only OA (fol-
low-up): 1 study, n =
17
Boldt PQoL (per- Post-intervention: 1 Post-intervention: No difference between groups.

2014 ceived quality study, n = 34, PQoL; 1
of life) study, n = 80, SQoL PQoL MD 10.8 (95% CI -4.2 to 25.8) (P =
0.16)
SQoL (subjec- Follow-up (interme-
tive quality of diate term): 1 study, SQoL MD 0.3 (95% CI -0.22 to 0.82) (P =
life) n = 80, SQoL 0.25)
Follow-up: SQoL MD 0.5 (95% CI -0.03

to 1.03) (P = 0.07)
Fransen QoL Post-intervention: 3 SMD 0.07 (95% CI -0.23 to 0.36) (ns) No difference between groups.
2014 studies, n = 183
Fransen QoL: self-report Post-intervention: 13 SMD 0.28 (95% CI 0.15 to 0.40) (P < Statistically significant, but
2015 questionnaire, studies, n = 1073 0.0001) equates to an absolute improve-
scale 0 to 100 ment of 4 points (95% CI 2 to 5)
(100 is maxi- Absolute difference 4% (95% CI 2% to on a 0 to 100 scale.
mum QoL) 5%)
Small effect size only.
relative difference 9% (95% CI 5% to
13%)
Gross QoL: SF-36 Post-intervention: 2 12-wk intervention: MD -2.22 (95% CI No significant difference between
2015a (Physical Func- studies, n = 143 -5.17 to 0.72) (ns) groups.
tion subscale)
24-wk intervention: MD 0.06 (95% CI
-4.06 to 4.17) (ns)
Lauret HRQoL No relevant studies n/a n/a

2014
Global assessment
Busch Global wellbe- Strength: 2 studies, n Strength: SMD 1.43 (95% CI 0.76 to Favoured exercise - higher score
2007 ing = 47 2.10) showed better QoL,
Aerobic: 4 studies, n Aerobic: SMD 0.49 (95% CI 0.23 to 0.75) Strength: very large effect size.
= 269
Aerobic: small-to-moderate ef-
fect size only.
Bidonde Participant-rat- 1 study, n = 46 MD -0.87 (95% CI -1.74 to 0.00) No effect.

2014 ed global (10-
cm VAS)
Gross Global per- 1 study, n = 70 "No significant difference" No significant difference.

2015a ceived effect
Collaboration.
Informed decisions.

Table 12. Quality of life (Continued)

Hayden Global assess- 7 studies, n = 16 Not reported n/a
2005 ment
Saragiotto Global impres- 1 study, n = 154 Short term, MD 1.30 (95% CI 0.30 to Medium effect size.
2016 sion of recovery 2.30) (P = 0.01)
Intermediate term, MD 1.20 (95% CI

0.31 to 2.09) (P = 0.008)
Long term, MD 1.50 (95% CI 0.61 to

2.39) (P < 0.001)
Yamato Global impres- 1 study, n = 86 Short term (< 3 months): MD 1.50 (95% "Low quality evidence (down-
2015 sion of recovery CI 0.70 to 2.30) graded due to imprecision and
inconsistency), we found a sig-
Intermediate term (3 to 12 months): nificant short-term effect, with a
MD 0.70 (95% CI -0.11 to 1.51) small effect size, but not for inter-
mediate/mid-term follow up."
Other method of assessment
Bidonde Multi-dimen- 7 studies, n = 367 MD -5.97 (95% CI -9.06 to -2.88) Favoured aquatic exercise - lower
2014 sional function- score showed reduced impact of
FIQ SMD -0.55 (95% CI -0.83 to -0.27) pain on life.
Absolute difference -6 (95% CI -9 to -3) "Moderate difference."
NNTB 5 (95% CI 3 to 9)
Busch Multi-dimen- 1 study, n = 60 SMD -1.27 (95% CI -1.83 to -0.72) Favoured exercise - lower score
2013 sional function showed reduced impact of pain
- FIQ Absolute difference -16.75 FIQ units on life.
(95% CI -23.31 to -10.19)
Very large effect size.
Hayden Work status 9 studies, n = 21 Not reported n/a

2005
Silva 2010 Health Assess- No included studies n/a n/a

ment Question-
naire (HAQ)
FIQ: Fibromyalgia Impact Questionnaire; HRQoL: health-related quality of life; MD: mean difference; n: number of participants; n/a: not
applicable; NNTB: number needed to treat for an additional beneficial outcome; OA: osteoarthritis; PQoL: perceived quality of life; QoL:
quality of life; SF-36: 36-item Short Form; SMD: standardised mean difference; SQoL: subjective quality of life; VAS: visual analogue scale.

Table 13. Adherence/withdrawals
Review Number of trials (and Number withdrawn (per Number RR or OR
participants) report- 1000) - intervention group withdrawn
ing withdrawals (per 1000)
- control
group
Bidonde 2014 8 studies, n = 472 151 (imputed from reported 129 (im- RR 1.13 (95% CI 0.73 to 1.77) (P
38/252) puted from = 0.45)
(fibromyalgia) reported
30/232)
Collaboration.
Informed decisions.

Table 13. Adherence/withdrawals (Continued)

Busch 2013 3 studies, n = 107 134 (95% CI 30 to 439) 39 RR 3.50 (95% CI 0.79 to 15.49)
(fibromyalgia)
Fransen 2014 7 studies, n = 715 59 (95% CI 30 to 114) 34 OR 1.77 (95% CI 0.86 to 3.65)
(osteoarthritis)
Han 2004 4 studies, n = 189 109 (imputed from reported 284 (im- RR 0.37 (95% CI 0.19 to 0.72)
11/101) puted from
(rheumatoid arthritis) reported
25/88)
Regnaux 2015 1 study, n = 102 44 (imputed from reported 0 Calculated RR 3.55 (95% CI
3/68 (4%); all from high-in- 0.19 to 66.8)
(osteoarthritis) tensity group)
Saragiotto 2016 7 studies, n = 671 0 0 -
(low back pain)
Silva 2010 No included studies n/a n/a n/a
(rheumatoid arthritis)
Total 30 studies, n = 2256 82.8/1000 81/1000 Calculated RR 1.02 (95% CI

0.94 to 1.12)
Calculated OR 1.05 (95% CI

0.88 to 1.25)
CI: confidence interval; n: number of participants; n/a: not applicable; OR: odds ratio; RR: risk ratio.

Table 14. Adverse events (not death)
Review Total Number Number Overall statement
number of trials of adverse
of trials (and par- events
(and participants)
ticipants) reporting
in review adverse
reporting events
exercise
vs control
in chronic
pain pop-
ulation
Bartels 4 (674) 2 (148) 0 Adverse events were recorded (and reported), but none occurred.
2007
Bidonde 9 (519) 0 0 Review stated that no included studies actively reported on adverse events
2014 (some reported withdrawal).
Boldt 3 (149) 2 (115) 5 events "Neck, shoulder and elbow injuries in five participants in the intervention
2014 over 2 stud- group."
ies
Collaboration.
Informed decisions.

Table 14. Adverse events (not death) (Continued)

Busch 34 (2276) 6 (strength Strength -
2007 training: training: 3
115, aero- Aerobic
bic: 1264) training: 5
Busch 3 (81) 2 (86 exer- 0 Adverse events were recorded (and reported), but none occurred.
2013 cising par-
ticipants)
Cramp 6 (371) 3 0 Adverse events were recorded (and reported), but none occurred.
2013
Fransen 10 (> 549) 5 7 events -

2014 over 3 stud-
ies
Fransen 54 (5362) 11 42 events -

2015 over 8 stud-
ies
Gross 16 (2485) 11 41 events -

2015a over 6 stud-
ies
Han 2004 3 (206) 2 1 event in 1 In narrative: "approximately one-third of the patients complained of sore-
study ness in the knee, shoulder or lower back during the first 3 weeks… pain
eventually subsided for all patients… only exception was one patient, who
complained of knee pain."
Hayden 43 (3907) 10 23 events "Negative reported: 16 events over 7 trials."

2005 over 10
studies
Hurkmans 5 (575) 2 0 Adverse events were recorded (and reported), but none occurred.
2009
Koopman 2 (68) 1 (10) 0 Adverse events were recorded (and reported), but none occurred.
2015
"The study investigated deleterious effects of this training on motor unit
survival through motor unit number estimates (MUNE). Results showed that
the MUNE did not change at the end of the training."
Lane 2014 30 (1822) 1 (88 exer- 2 events RR 0.20 (95% CI 0.01 to 4.15) in favour of exercise group.
cising par- in control
ticipants) group in 1
study
Regnaux 1 (102) 1 (68 ex- 3 events in "3 participants in high resistance group discontinued the exercise interven-
2015 ercising 1 study tion due to severe knee pain."
partici-
pants over
2 groups:
low and
high resis-
tance)
Collaboration.
Informed decisions.

Table 14. Adverse events (not death) (Continued)

Saragiotto 7 (671) 1 (154) 5 events in "Five patients (three from the MCE [motor control exercise] group and two
2016 1 study from the minimal intervention group) had mild adverse effects during the
study (all temporary exacerbations of pain)."
van der 10 (1690) 0 0 Of the relevant studies, none actively reported on adverse events.
Heijden
2015
Yamato 6 (265) 1 (86) 0 Adverse events were recorded (and reported), but none occurred.
2015
Total 246 stud- 61 studies 137 events 61/246 (25%) of studies have reported on adverse events; of which
ies over 39 39/61 (64%) did have adverse events occur as a result of the interven-
(> 2134 studies tion or control.
(> 21,772) partici-
pants)
n: number of participants; RR: risk ratio.

WHAT'S NEW

Date Event Description
18 February 2020 Amended Clarification added to Declarations of interest.

HISTORY
Protocol first published: Issue 8, 2014
Review first published: Issue 1, 2017

Date Event Description
9 November 2018 Amended "Next stage expected" date extended to 2022; we assess all
overviews for updating five years after publication.
18 April 2017 New citation required but conclusions Conclusions not changed; retrospective open access.
have not changed
10 April 2017 Amended See Published notes.

CONTRIBUTIONS OF AUTHORS
LG conceived the idea of the overview, and wrote the protocol and full overview.
LG, BHS, LC, and RAM developed the concept and details of the overview (participants, intervention, comparison, outcomes).
LG and CC carried out searches and selected reviews for inclusion (RAM and DM acted as arbitrators).
LG and CC carried out assessment of methodological quality using the AMSTAR tool (DM acted as arbitrator).
LG and RAM extracted data and interpreted initial findings.
LG, RAM, LC, and BHS formulated the focus of the discussion and made suggestions for future study and review authors.
Collaboration.
Informed decisions.

All authors were involved in the interpretation of results, and in approving the final review.
LG and BHS will be responsible for future updates.
DECLARATIONS OF INTEREST
LG: none known.
RAM has received grant support from Grünenthal relating to individual patient level analyses of trial data regarding tapentadol in
osteoarthritis and back pain (2015). He has received honoraria for attending boards with Menarini concerning methods of analgesic trial
design (2014), with Novartis (2014) about the design of network meta-analyses, and RB on understanding pharmacokinetics of drug uptake
(2015). He has received honoraria from Omega Pharma (2016) and Futura Pharma (2016) for providing advice on trial and data analysis
methods.
CC: none known.
DM: none known.
LC: none known. For transparency, LC has received honoraria for speaking at educational meetings to healthcare professionals on a range
of chronic pain topics (Pfizer (October 2015), Astellas (June 2014, March 2015)); editor on the British Journal of Anaesthesia (receives an
honorarium plus a contribution toward related departmental expenses (October 2010 - to date)). LC is a medical clinician attending patients
in the NHS Lothian Pain Service.
BHS: none known. BHS is a medical clinician attending patients in the NHS Tayside Pain Service.
This review was identified in a 2019 audit as not meeting the current definition of the Cochrane Commercial Sponsorship policy. At the
time of its publication it was compliant with the interpretation of the existing policy. As with all reviews, new and updated, at update this
review will be revised according to 2020 policy update.
SOURCES OF SUPPORT
Internal sources
• No sources of support supplied
External sources
• MRC, Lifelong Health and Wellbeing Grant, ID 91029, UK.
• National Institute of Academic Anaesthesia (BJA RCoA NIAA), UK.
NOTES
This overview review was re-published in April 2017 with retrospective open access.
INDEX TERMS
Medical Subject Headings (MeSH)

Chronic Pain [mortality] [psychology] [*therapy]; Exercise Therapy [adverse effects] [*methods]; Health Services Needs and Demand;
Myalgia [etiology]; Pain Measurement; Patient Compliance; Quality of Life; Randomized Controlled Trials as Topic; Review Literature
as Topic
MeSH check words

Adult; Humans
Collaboration.

Cochrane: Library

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Cochrane: Library

Uploaded by

Copyright:

Available Formats

Geneen LJ, Moore RA, Clarke C, Martin D, Colvin LA, Smith BH

Geneen LJ, Moore RA, Clarke C, Martin D, Colvin LA, Smith BH.

Physical activity and exercise for chronic pain in adults: an overview of

Contact address: Louise J Geneen, Oxford, UK. louisegeneen@gmail.com.

Editorial group: Cochrane Pain, Palliative and Supportive Care Group

Healthcare use/attendance: was not reported in any review.

Key results and quality of the evidence

BACKGROUND recommendations on the use of exercise, based on evidence drawn

endogenous opioids, leading to transient anti-nociception in both OBJECTIVES

Figure 1. Study flow diagram.

Yoga week × number of weeks), for a more accurate and detailed

HAQ 1.05/3 (1 study, n = 249)

11.05 to 22.6 on a 0 to 150 WUSPI VAS 35/100 to 61/100 44/100

Mean pain score 46/100 (95% CI 41 to - 40/100 to 60/100

(1 study, n = 55; Koopman 2015)

- - range 25/100 to 56/100

(4 studies, n = 291; Saragiotto 2016)

(2 studies, n = 41; van der Heijden 2015)

- - range 18/100 to 52/100

(6 studies, n = 148; Yamato 2015)

Range: 46 to 70.9 on a 0 to 100 scale Range: 35 to 55 on a 0 to 100 Range: 18 to 60 on a 0 to 100 scale

Treatment effect Treatment effect

Saragiotto 2016 Craane 2006

Gross 2015b Jordan 2010

Henschke 2010 Karjalainen 2003

Hurley 2013 Miller 2014

O'Connell 2013 Database of Systematic Reviews 2012, Issue 1. [DOI:

Østerås 2013 Collins 1997

Exercise is Medicine Messier 2013

February 2014. www.nice.org.uk/guidance/cg177 (accessed 4 SIGN 2013

Ursini 2011 Waddell 1987

van Hecke 2013a WHO 2015

van Hecke 2013b Wieland 2013

Vigorito 2014 Williams 2012

Table 1. AMSTAR tool to assess the methodological quality of systematic reviews (Continued)

Aggarwal 2011 Not exercise/physical activity

Table 2. Reasons for exclusion (Continued)

Brønfort 2014 Withdrawn from the Cochrane Library

Choi 2010 Not chronic using definition of > 3 months

Dagfinrud 2008 Physiotherapy - required therapist to perform intervention

Dahm 2010 Acute pain, not chronic. Intervention was advice

Dal Bello-Haas 2013 Malignant condition

de Souza 2012 Drug- and surgery-based interventions

Franke 2015 Not exercise/physical activity

Green 2003 Physiotherapy - required therapist to perform intervention

Gross 1998 Withdrawn from the Cochrane Library

Gross 2012 Not exercise/physical activity

Gross 2015b Not exercise/physical activity

Heintjes 2003 Withdrawn from the Cochrane Library January 2015

Henschke 2010 Not exercise/physical activity

Heymans 2004 Exercise could not be assessed as stand-alone intervention

Hilde 2006 Withdrawn from the Cochrane Library

Hoving 2014 No exercise intervention, and no pain outcome measure

Jones 2000 Drug-based interventions

Jordan 2010 Intervention to improve adherence to exercise, not exercise itself

Kamper 2014 Exercise could not be assessed as stand-alone intervention

Karjalainen 1999 Exercise could not be assessed as stand-alone intervention

Karjalainen 2003 Exercise could not be assessed as stand-alone intervention

Table 2. Reasons for exclusion (Continued)

Liddle 2015 Pain in pregnancy only, not chronic pain

Moi 2013 Exercise could not be assessed as stand-alone intervention

O'Brien 2004 No pain outcome measure

O'Connell 2013 Overview of reviews, not systematic review